The fastest, litest AI Gateway. Rust core with Python SDK. Call 100+ LLM APIs in OpenAI (or native) format with cost tracking, guardrails, load balancing, and logging [Bedrock, Azure, OpenAI, Anthropic, OpenAI, VertexAI, vLLM, Nvidia NIM]
Python ai-gatewayanthropicazure-openaibedrockgatewaylangchainlitellmllmllm-gatewayllmopsmcp-gatewayopenaiopenai-proxyrustrust-aivertex-ai
/**
* This file was auto-generated by openapi-typescript.
* Do not make direct changes to the file.
*/
export interface paths {
"/": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Home */
get: operations["home__get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/.well-known/agent-skills/index.json": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Agent Skills Index
* @description Agent Skills v0.2.0 discovery index over every skill stored on this proxy.
*/
get: operations["agent_skills_index__well_known_agent_skills_index_json_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/.well-known/jwks.json": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Jwks Json
* @description JSON Web Key Set endpoint.
*
* Returns the RSA public key used by MCPJWTSigner to sign outbound MCP tokens.
* MCP servers and gateways use this endpoint to verify liteLLM-issued JWTs.
*
* Returns an empty key set if MCPJWTSigner is not configured.
*/
get: operations["jwks_json__well_known_jwks_json_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/.well-known/litellm-cli-auth": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Native Client Auth Discovery
* @description The versioned contract a native client (``lite login --pkce``, or a CLI in any other
* language) reads to sign a user in through the browser and obtain a proxy credential.
*/
get: operations["native_client_auth_discovery__well_known_litellm_cli_auth_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/.well-known/litellm-ui-config": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get Ui Config */
get: operations["get_ui_config__well_known_litellm_ui_config_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/.well-known/oauth-authorization-server": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Oauth Authorization Server Mcp
* @description OAuth authorization server discovery endpoint.
*
* Supports both legacy pattern (/{server_name}) and root endpoint.
*/
get: operations["oauth_authorization_server_mcp__well_known_oauth_authorization_server_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/.well-known/oauth-authorization-server/mcp": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Oauth Authorization Server Aggregate
* @description OAuth authorization server discovery for the aggregate /mcp endpoint, the RFC 8414
* path-inserted form for a client that treats {base}/mcp as its authorization base URL.
*
* The single-segment /mcp is reserved for the aggregate so the discovery chain stays
* consistent: the aggregate protected-resource document advertises {base}/mcp as its
* authorization server, so the document served here must have issuer {base}/mcp. A server
* literally named ``mcp`` therefore does not take this route; it keeps its standard
* two-segment discovery at /.well-known/oauth-authorization-server/mcp/mcp. Letting the
* per-server row win here instead would serve an issuer of {base} against a resource that
* advertised {base}/mcp, which fails the RFC 8414 issuer check and breaks the front door.
*/
get: operations["oauth_authorization_server_aggregate__well_known_oauth_authorization_server_mcp_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/.well-known/oauth-authorization-server/mcp/{mcp_server_name}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Oauth Authorization Server Mcp Standard
* @description OAuth authorization server discovery endpoint using standard MCP URL pattern.
*
* Standard pattern: /mcp/{server_name}
* Discovery path: /.well-known/oauth-authorization-server/mcp/{server_name}
*/
get: operations["oauth_authorization_server_mcp_standard__well_known_oauth_authorization_server_mcp__mcp_server_name__get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/.well-known/oauth-authorization-server/{mcp_server_name}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Oauth Authorization Server Mcp
* @description OAuth authorization server discovery endpoint.
*
* Supports both legacy pattern (/{server_name}) and root endpoint.
*/
get: operations["oauth_authorization_server_mcp__well_known_oauth_authorization_server__mcp_server_name__get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/.well-known/oauth-authorization-server/{mcp_server_name}/mcp": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Oauth Authorization Server Legacy
* @description OAuth authorization server discovery for legacy /{server_name}/mcp pattern.
*/
get: operations["oauth_authorization_server_legacy__well_known_oauth_authorization_server__mcp_server_name__mcp_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/.well-known/oauth-protected-resource": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Oauth Protected Resource Root */
get: operations["oauth_protected_resource_root__well_known_oauth_protected_resource_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/.well-known/oauth-protected-resource/mcp": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Oauth Protected Resource Aggregate
* @description OAuth protected resource discovery for the aggregate /mcp endpoint.
*
* The single-segment ``/mcp`` path does not collide with any per-server PRM pattern
* (those are two-segment: ``/mcp/{server}`` or ``/{server}/mcp``), so this unambiguously
* describes the aggregate resource.
*/
get: operations["oauth_protected_resource_aggregate__well_known_oauth_protected_resource_mcp_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/.well-known/oauth-protected-resource/mcp/{mcp_server_name}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Oauth Protected Resource Mcp Standard
* @description OAuth protected resource discovery endpoint using standard MCP URL pattern.
*
* Standard pattern: /mcp/{server_name}
* Discovery path: /.well-known/oauth-protected-resource/mcp/{server_name}
*
* This endpoint is compliant with MCP specification and works with standard
* MCP clients like mcp-inspector and VSCode Copilot.
*/
get: operations["oauth_protected_resource_mcp_standard__well_known_oauth_protected_resource_mcp__mcp_server_name__get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/.well-known/oauth-protected-resource/{mcp_server_name}/mcp": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Oauth Protected Resource Mcp
* @description OAuth protected resource discovery endpoint using LiteLLM legacy URL pattern.
*
* Legacy pattern: /{server_name}/mcp
* Discovery path: /.well-known/oauth-protected-resource/{server_name}/mcp
*
* This endpoint is kept for backward compatibility. New integrations should
* use the standard MCP pattern (/mcp/{server_name}) instead.
*/
get: operations["oauth_protected_resource_mcp__well_known_oauth_protected_resource__mcp_server_name__mcp_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/.well-known/openid-configuration": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Openid Configuration */
get: operations["openid_configuration__well_known_openid_configuration_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/.well-known/skills/index.json": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Agent Skills Index
* @description Agent Skills v0.2.0 discovery index over every skill stored on this proxy.
*/
get: operations["agent_skills_index__well_known_skills_index_json_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/a2a/{agent_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Invoke Agent A2A
* @description Invoke an agent using the A2A protocol (JSON-RPC 2.0).
*
* Supported methods:
* - message/send: Send a message and get a response
* - message/stream: Send a message and stream the response
*/
post: operations["invoke_agent_a2a_a2a__agent_id__post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/a2a/{agent_id}/.well-known/agent-card.json": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Agent Card
* @description Get the agent card for an agent (A2A discovery endpoint).
*
* Supports both standard paths:
* - /.well-known/agent-card.json
* - /.well-known/agent.json
*
* The URL in the agent card is rewritten to point to the LiteLLM proxy,
* so all subsequent A2A calls go through LiteLLM for logging and cost tracking.
*/
get: operations["get_agent_card_a2a__agent_id___well_known_agent_card_json_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/a2a/{agent_id}/.well-known/agent.json": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Agent Card
* @description Get the agent card for an agent (A2A discovery endpoint).
*
* Supports both standard paths:
* - /.well-known/agent-card.json
* - /.well-known/agent.json
*
* The URL in the agent card is rewritten to point to the LiteLLM proxy,
* so all subsequent A2A calls go through LiteLLM for logging and cost tracking.
*/
get: operations["get_agent_card_a2a__agent_id___well_known_agent_json_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/a2a/{agent_id}/message/send": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Invoke Agent A2A
* @description Invoke an agent using the A2A protocol (JSON-RPC 2.0).
*
* Supported methods:
* - message/send: Send a message and get a response
* - message/stream: Send a message and stream the response
*/
post: operations["invoke_agent_a2a_a2a__agent_id__message_send_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/access_group/list": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List Access Groups
* @description List all access groups.
*
* Returns a list of all access groups with their model names, deployment counts, shared budget
* and the spend drawn against it.
*
* Example:
* ```bash
* curl -X GET 'http://localhost:4000/access_group/list' \
* -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*
* Returns:
* - ListAccessGroupsResponse with all access groups
*/
get: operations["list_access_groups_access_group_list_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/access_group/new": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Create Model Group
* @description Create a new access group containing multiple model names.
*
* An access group is a named collection of model groups that can be referenced
* by teams/keys for simplified access control.
*
* Example:
* ```bash
* curl -X POST 'http://localhost:4000/access_group/new' \
* -H "Authorization: Bearer $LITELLM_MASTER_KEY" \
* -H 'Content-Type: application/json' \
* -d '{
* "access_group": "production-models",
* "model_names": ["gpt-4", "claude-3-opus", "gemini-pro"]
* }'
* ```
*
* Parameters:
* - access_group: str - The access group name (e.g., "production-models")
* - model_names: List[str] - List of existing model groups to include
*
* Returns:
* - NewModelGroupResponse with the created access group details
*
* Raises:
* - HTTPException 400: If any model names don't exist
* - HTTPException 500: If database operations fail
*/
post: operations["create_model_group_access_group_new_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/access_group/{access_group}/budget": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Access Group Budget
* @description Get the shared budget of an access group, and the spend drawn against it.
*
* Example:
* ```bash
* curl -X GET 'http://localhost:4000/access_group/production-models/budget' \
* -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*
* Parameters:
* - access_group: str - The access group name (URL path parameter)
*
* Returns:
* - AccessGroupBudgetResponse; budget is null when the group has no budget set
*
* Raises:
* - HTTPException 404: If access group not found
*/
get: operations["get_access_group_budget_access_group__access_group__budget_get"];
/**
* Set Access Group Budget
* @description Set or replace the shared budget of an access group. Idempotent.
*
* Every key that can reach a model in the group draws from this one budget.
*
* Example:
* ```bash
* curl -X PUT 'http://localhost:4000/access_group/production-models/budget' \
* -H "Authorization: Bearer $LITELLM_MASTER_KEY" \
* -H 'Content-Type: application/json' \
* -d '{
* "max_budget": 100.0,
* "budget_duration": "30d"
* }'
* ```
*
* Parameters:
* - access_group: str - The access group name (URL path parameter)
* - max_budget: Optional[float] - Requests fail once the group's shared spend exceeds this
* - soft_budget: Optional[float] - Fires an alert when reached; requests still succeed
* - budget_duration: Optional[str] - Frequency of resetting the group's spend (e.g. '30d')
* - budget_id: Optional[str] - Link an existing budget instead of creating one
*
* Returns:
* - AccessGroupBudgetResponse with the stored budget and current spend
*
* Raises:
* - HTTPException 400: If no budget field is given, or budget_duration cannot be parsed
* - HTTPException 404: If access group not found
*/
put: operations["set_access_group_budget_access_group__access_group__budget_put"];
post?: never;
/**
* Delete Access Group Budget
* @description Clear the shared budget of an access group, leaving the group itself in place.
*
* Example:
* ```bash
* curl -X DELETE 'http://localhost:4000/access_group/production-models/budget' \
* -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*
* Parameters:
* - access_group: str - The access group name (URL path parameter)
*
* Returns:
* - DeleteAccessGroupBudgetResponse; budget_deleted is false when there was nothing to clear
*
* Raises:
* - HTTPException 404: If access group not found
*/
delete: operations["delete_access_group_budget_access_group__access_group__budget_delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/access_group/{access_group}/delete": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
post?: never;
/**
* Delete Access Group
* @description Delete an access group.
*
* Removes the access group from all deployments that have it.
*
* Example:
* ```bash
* curl -X DELETE 'http://localhost:4000/access_group/production-models/delete' \
* -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*
* Parameters:
* - access_group: str - The access group name (URL path parameter)
*
* Returns:
* - DeleteModelGroupResponse with deletion details
*
* Raises:
* - HTTPException 404: If access group not found
*/
delete: operations["delete_access_group_access_group__access_group__delete_delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/access_group/{access_group}/info": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Access Group Info
* @description Get information about a specific access group.
*
* Example:
* ```bash
* curl -X GET 'http://localhost:4000/access_group/production-models/info' \
* -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*
* Parameters:
* - access_group: str - The access group name (URL path parameter)
*
* Returns:
* - AccessGroupInfo with the access group details, its shared budget and its spend
*
* Raises:
* - HTTPException 404: If access group not found
*/
get: operations["get_access_group_info_access_group__access_group__info_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/access_group/{access_group}/update": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
/**
* Update Access Group
* @description Update an access group's model names.
*
* This will:
* 1. Remove the access group from all current deployments
* 2. Add the access group to all deployments for the new model_names list
*
* Example:
* ```bash
* curl -X PUT 'http://localhost:4000/access_group/production-models/update' \
* -H "Authorization: Bearer $LITELLM_MASTER_KEY" \
* -H 'Content-Type: application/json' \
* -d '{
* "model_names": ["gpt-4", "claude-3-sonnet"]
* }'
* ```
*
* Parameters:
* - access_group: str - The access group name (URL path parameter)
* - model_names: List[str] - New list of model groups to include
*
* Returns:
* - NewModelGroupResponse with the updated access group details
*
* Raises:
* - HTTPException 400: If any model names don't exist
* - HTTPException 404: If access group not found
*/
put: operations["update_access_group_access_group__access_group__update_put"];
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/active/callbacks": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Active Callbacks
* @description Returns a list of litellm level settings
*
* This is useful for debugging and ensuring the proxy server is configured correctly.
*
* Response schema:
* ```
* {
* "alerting": _alerting,
* "litellm.callbacks": litellm_callbacks,
* "litellm.input_callback": litellm_input_callbacks,
* "litellm.failure_callback": litellm_failure_callbacks,
* "litellm.success_callback": litellm_success_callbacks,
* "litellm._async_success_callback": litellm_async_success_callbacks,
* "litellm._async_failure_callback": litellm_async_failure_callbacks,
* "litellm._async_input_callback": litellm_async_input_callbacks,
* "all_litellm_callbacks": all_litellm_callbacks,
* "num_callbacks": len(all_litellm_callbacks),
* "num_alerting": _num_alerting,
* "litellm.request_timeout": litellm.request_timeout,
* }
* ```
*/
get: operations["active_callbacks_active_callbacks_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/adaptive_router/state": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Adaptive Router State
* @description Return live bandit posteriors + queue depth for every configured adaptive router.
*
* Admin-only. Returns 404 if no adaptive router is configured.
*
* Response shape: `{"routers": [<snapshot>, ...]}` — one snapshot per
* adaptive-router deployment. Each snapshot's `router_name` field identifies
* which deployment it came from.
*/
get: operations["get_adaptive_router_state_adaptive_router_state_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/add/allowed_ip": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/** Add Allowed Ip */
post: operations["add_allowed_ip_add_allowed_ip_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/agent/daily/activity": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Agent Daily Activity
* @description Get daily activity for specific agents or all accessible agents.
*/
get: operations["get_agent_daily_activity_agent_daily_activity_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/agent/daily/activity/aggregated": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get Agent Daily Activity Aggregated */
get: operations["get_agent_daily_activity_aggregated_agent_daily_activity_aggregated_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/agent/daily/activity/aggregated/keys": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get Agent Daily Activity Aggregated Keys */
get: operations["get_agent_daily_activity_aggregated_keys_agent_daily_activity_aggregated_keys_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/agent/daily/activity/aggregated/model_top_keys": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get Agent Daily Activity Model Top Keys */
get: operations["get_agent_daily_activity_model_top_keys_agent_daily_activity_aggregated_model_top_keys_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/agent/daily/activity/aggregated/search": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get Agent Daily Activity Aggregated Search */
get: operations["get_agent_daily_activity_aggregated_search_agent_daily_activity_aggregated_search_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/agent/daily/activity/export": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get Agent Daily Activity Export */
get: operations["get_agent_daily_activity_export_agent_daily_activity_export_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/alerting/settings": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Alerting Settings
* @description Return the configurable alerting param, description, and current value
*/
get: operations["alerting_settings_alerting_settings_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/anthropic/{endpoint}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Anthropic Proxy Route
* @description [Docs](https://docs.litellm.ai/docs/pass_through/anthropic_completion)
*/
get: operations["anthropic_proxy_route_anthropic__endpoint__get"];
/**
* Anthropic Proxy Route
* @description [Docs](https://docs.litellm.ai/docs/pass_through/anthropic_completion)
*/
put: operations["anthropic_proxy_route_anthropic__endpoint__put"];
/**
* Anthropic Proxy Route
* @description [Docs](https://docs.litellm.ai/docs/pass_through/anthropic_completion)
*/
post: operations["anthropic_proxy_route_anthropic__endpoint__post"];
/**
* Anthropic Proxy Route
* @description [Docs](https://docs.litellm.ai/docs/pass_through/anthropic_completion)
*/
delete: operations["anthropic_proxy_route_anthropic__endpoint__delete"];
options?: never;
head?: never;
/**
* Anthropic Proxy Route
* @description [Docs](https://docs.litellm.ai/docs/pass_through/anthropic_completion)
*/
patch: operations["anthropic_proxy_route_anthropic__endpoint__patch"];
trace?: never;
};
"/api/event_logging/batch": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Event Logging Batch
* @description Stubbed endpoint for Anthropic event logging batch requests.
*
* This endpoint accepts event logging requests but does nothing with them.
* It exists to prevent 404 errors from Claude Code clients that send telemetry.
*/
post: operations["event_logging_batch_api_event_logging_batch_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/api/plugins": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List Plugins
* @description Return registered plugins for authenticated UI callers.
*
* plugin_key is never returned — the browser never needs it (the proxy injects
* it server-side from the registry), and exposing it here would leak the
* credential into React state and DevTools. Admin key management goes through
* the redacted /config/field/info path instead.
*/
get: operations["list_plugins_api_plugins_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/api/plugins/auth-token": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Plugin Auth Token
* @description Issue a short-lived, audience-scoped plugin session claim.
*
* The claim contains {user_id, user_role, plugin, exp}. It does NOT
* contain the caller's litellm bearer token — a compromised plugin can
* only learn the caller's identity, not impersonate them against the proxy.
*
* Encrypted with a key derived from HMAC(LITELLM_SALT_KEY, plugin_name),
* so each plugin holds only its own key and cannot forge claims for others.
*
* Requires LITELLM_SALT_KEY to be set; returns 503 otherwise.
*/
get: operations["plugin_auth_token_api_plugins_auth_token_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/apply_guardrail": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Apply Guardrail
* @description Apply a guardrail to text input and return the processed result.
*
* This endpoint allows testing guardrails by applying them to custom text inputs.
*/
post: operations["apply_guardrail_apply_guardrail_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/assemblyai/{endpoint}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Assemblyai Proxy Route */
get: operations["assemblyai_proxy_route_assemblyai__endpoint__get"];
/** Assemblyai Proxy Route */
put: operations["assemblyai_proxy_route_assemblyai__endpoint__put"];
/** Assemblyai Proxy Route */
post: operations["assemblyai_proxy_route_assemblyai__endpoint__post"];
/** Assemblyai Proxy Route */
delete: operations["assemblyai_proxy_route_assemblyai__endpoint__delete"];
options?: never;
head?: never;
/** Assemblyai Proxy Route */
patch: operations["assemblyai_proxy_route_assemblyai__endpoint__patch"];
trace?: never;
};
"/assistants": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Assistants
* @description Returns a list of assistants.
*
* API Reference docs - https://platform.openai.com/docs/api-reference/assistants/listAssistants
*/
get: operations["get_assistants_assistants_get"];
put?: never;
/**
* Create Assistant
* @description Create assistant
*
* API Reference docs - https://platform.openai.com/docs/api-reference/assistants/createAssistant
*/
post: operations["create_assistant_assistants_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/assistants/{assistant_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
post?: never;
/**
* Delete Assistant
* @description Delete assistant
*
* API Reference docs - https://platform.openai.com/docs/api-reference/assistants/createAssistant
*/
delete: operations["delete_assistant_assistants__assistant_id__delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/audio/speech": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Audio Speech
* @description Same params as:
*
* https://platform.openai.com/docs/api-reference/audio/createSpeech
*/
post: operations["audio_speech_audio_speech_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/audio/transcriptions": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Audio Transcriptions
* @description Same params as:
*
* https://platform.openai.com/docs/api-reference/audio/createTranscription?lang=curl
*/
post: operations["audio_transcriptions_audio_transcriptions_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/audit": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Audit Logs
* @description Get all audit logs with filtering and pagination.
*
* Returns a paginated response of audit logs matching the specified filters.
*
* Note: object_team_id and object_key_hash use Prisma JSON path filtering,
* which requires PostgreSQL.
*/
get: operations["get_audit_logs_audit_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/audit/{id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Audit Log By Id
* @description Get detailed information about a specific audit log entry by its ID.
*
* Args:
* id (str): The unique identifier of the audit log entry
*
* Returns:
* AuditLogResponse: Detailed information about the audit log entry
*
* Raises:
* HTTPException: If the audit log is not found or if there's a database connection error
*/
get: operations["get_audit_log_by_id_audit__id__get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/authorize": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Authorize */
get: operations["authorize_authorize_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/authorize/complete": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Authorize Complete
* @description Finish an aggregate connect flow: mint the gateway authorization code for the
* signed-in user and hand it back to the DCR client, by 303 redirect (default) or, for
* a loopback client on a different machine, as a copyable callback URL
* (``delivery=manual``). POST plus the per-flow HttpOnly cookie set at /authorize; an
* anonymous or bad-flow request just 400s. The native-client consent page adds
* ``decision`` (approve or deny) and the ``team_id`` the credential is attributed to.
*/
post: operations["authorize_complete_authorize_complete_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/authorize/flow": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Authorize Flow */
get: operations["authorize_flow_authorize_flow_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/authorize/mcp-session": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Authorize Mcp Session */
get: operations["authorize_mcp_session_authorize_mcp_session_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/auto_router/availability": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/** Get Auto Router Availability */
post: operations["get_auto_router_availability_auto_router_availability_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/auto_router/benchmarks": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Auto Router Benchmarks
* @description Benchmarks for the auto-router dashboard: session shape, savings against the configured
* baseline, and prompt-caching behaviour bucketed by what the router did.
*
* Reads session rollups folded once per request at spend-write time, so this endpoint
* never scans LiteLLM_SpendLogs. A user filter selects only turns attributed to that
* internal user when written; older key-only history remains outside user views. Money counts
* only requests on the selected UTC days, and the all-router savings headline is the same daily
* total the Overall view reads. Session shape and caching cover every session that overlaps the
* window, whole. Overall hit rate is over telemetry-bearing turns; each bucket's hit rate is
* over that bucket's turns.
*
* The rollup supplies the measures, never the list. Which routers appear comes from the
* model registry, so one shows up as soon as it is configured and reads zero until it
* serves traffic, and `routers_in_scope` counts those too rather than only the routers the
* window recorded.
*/
get: operations["get_auto_router_benchmarks_auto_router_benchmarks_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/auto_router/classifier/default_prompt": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Auto Router Classifier Default Prompt
* @description Get the built-in system prompt used by an auto-router's LLM classifier
*/
get: operations["get_auto_router_classifier_default_prompt_auto_router_classifier_default_prompt_get"];
put?: never;
/**
* Preview Auto Router Classifier Prompt
* @description Get the system prompt an auto-router's LLM classifier sends for an edited tier set
*/
post: operations["preview_auto_router_classifier_prompt_auto_router_classifier_default_prompt_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/auto_router/session": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Auto Router Session
* @description One auto-routed session, for the key that ran it: the model its last turn was routed to and the
* session's spend against the router's savings baseline. Built for a coding agent's status line
* or stop hook, so any virtual key may call it and only ever sees rows written under its own
* key hash. Reads the LiteLLM_AutoRouterSession rollup, which the asynchronous spend flush
* fills a moment after each turn; a session with no flushed auto-routed turn yet is a 404. The
* id is bounded the way the writer bounded it, so an oversized client id still finds its row.
*/
get: operations["get_auto_router_session_auto_router_session_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/auto_router/shadow_eval": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List Shadow Eval Jobs
* @description List shadow eval jobs, newest first, each target with its attempt count so status
* is accurate. Judged counts, spend, and results ride the detail endpoint only.
*/
get: operations["list_shadow_eval_jobs_auto_router_shadow_eval_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/auto_router/shadow_eval/start": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Start Shadow Eval
* @description Start a shadow eval: duplicate a sampled slice of one or more targets' live traffic
* against a second arm, judge the two responses blind, and stratify win rates by tier,
* by the model that served the real arm, and by target.
*
* A target is a virtual key, a team, or a user. Team and user targets match on the
* identity every request resolves to at auth time, so they cover JWT-authenticated
* traffic, which presents no virtual key; a user target samples that user's traffic
* across all their teams, whether it arrives on a JWT or a key they own. models narrows
* every target to requests for those model groups, so a user plus one model samples that
* user's traffic on that model across every key they own; it is forward-only, since a
* reverse job already samples exactly the traffic its own router served.
*
* A forward job answers whether the targets should adopt router_name: it samples the
* requests the router did not serve and duplicates them through it. A reverse job
* answers whether a target already on the router still gains from it: it samples the
* requests the router did serve and duplicates them against baseline_model. A target
* can hold one active job per direction, so both questions can run at once, and a
* request matching several jobs' targets (say its key and its team) is sampled by
* each, separately budgeted.
*
* Shadow responses are never served to users. Each target samples until its recorded
* eval spend, the shadow and judge calls' own cost, reaches max_budget dollars, the
* job's window ends, or the job is stopped, so one target running out of budget does
* not end sampling for the others; sampling changes propagate to pods within about 10
* seconds. Shadow and judge calls bill to the sampled request's own identity but are
* excluded from request counts and auto-router adoption metrics.
*/
post: operations["start_shadow_eval_auto_router_shadow_eval_start_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/auto_router/shadow_eval/{job_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Shadow Eval Job
* @description One job with derived counts, judge spend, latest error, and stratified results.
*/
get: operations["get_shadow_eval_job_auto_router_shadow_eval__job_id__get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/auto_router/shadow_eval/{job_id}/stop": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Stop Shadow Eval Job
* @description Stop an active shadow eval job, every target it scopes at once. Attempts are kept;
* sampling halts within ~10s. Targets that already stopped on their own budget keep the
* stopped_at they earned. The statement is the whole state machine: it claims the job
* only while a leg still samples inside the window with no stop recorded, so a racing
* operator, a same-instant budget spend, and a repeat stop all read the same 400 with
* the status the job actually holds.
*/
post: operations["stop_shadow_eval_job_auto_router_shadow_eval__job_id__stop_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/auto_router/test_routing": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Preview Auto Router Routing
* @description Route a single request through a complexity-router config and report where it landed.
*
* Answers "which model would this request get?" for a config that only exists in a form,
* so an auto router can be checked before it is created. The request is classified by the
* same pre-routing hook a live request runs, over the same messages, system prompt and tool
* definitions, then dropped: nothing is sent to the model it routed to, and no auto router is
* created. A heuristic config therefore spends nothing, while an `llm` classifier or semantic
* keyword matching bills its classifier/embedding call to the calling key, like Test Connection
* does.
*
* Send `messages` to classify a real turn, with `system` and `tools` beside it when the surface
* carries them top level, as Anthropic /v1/messages does. `prompt` is the single-ask shorthand and
* routes as one user turn with nothing around it.
*
* **Example Request:**
* ```json
* {
* "messages": [
* {"role": "system", "content": "You are a database migration assistant"},
* {"role": "user", "content": "the index is not unique"},
* {"role": "assistant", "content": "Then two workers can both insert. Add a unique index"},
* {"role": "user", "content": "ok do it"}
* ],
* "tools": [{"type": "function", "function": {"name": "Bash", "description": "Run a command"}}],
* "complexity_router_config": {
* "tiers": {"SIMPLE": ["gpt-4o-mini"], "REASONING": ["o3"]},
* "classifier_type": "heuristic"
* }
* }
* ```
*/
post: operations["preview_auto_router_routing_auto_router_test_routing_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/auto_router/validate_complexity_router_config": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Validate Complexity Router Config
* @description Validate a complexity-router config without saving it.
*
* Runs the same check every write path runs (the router's own pydantic model), so a form can
* show the backend's exact verdict while the operator is still editing rather than after a
* rejected save. Uses the same team opt-in and model-access checks as configuration
* writes for members. Nothing is created, routed, or billed.
*/
post: operations["validate_complexity_router_config_auto_router_validate_complexity_router_config_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/azure/{endpoint}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Azure Proxy Route
* @description Call any azure endpoint using the proxy.
*
* Just use `{PROXY_BASE_URL}/azure/{endpoint:path}`
*
* Checks if the deployment id in the url is a litellm model name. If so, it will route using the llm_router.allm_passthrough_route.
*/
get: operations["azure_proxy_route_azure__endpoint__get"];
/**
* Azure Proxy Route
* @description Call any azure endpoint using the proxy.
*
* Just use `{PROXY_BASE_URL}/azure/{endpoint:path}`
*
* Checks if the deployment id in the url is a litellm model name. If so, it will route using the llm_router.allm_passthrough_route.
*/
put: operations["azure_proxy_route_azure__endpoint__put"];
/**
* Azure Proxy Route
* @description Call any azure endpoint using the proxy.
*
* Just use `{PROXY_BASE_URL}/azure/{endpoint:path}`
*
* Checks if the deployment id in the url is a litellm model name. If so, it will route using the llm_router.allm_passthrough_route.
*/
post: operations["azure_proxy_route_azure__endpoint__post"];
/**
* Azure Proxy Route
* @description Call any azure endpoint using the proxy.
*
* Just use `{PROXY_BASE_URL}/azure/{endpoint:path}`
*
* Checks if the deployment id in the url is a litellm model name. If so, it will route using the llm_router.allm_passthrough_route.
*/
delete: operations["azure_proxy_route_azure__endpoint__delete"];
options?: never;
head?: never;
/**
* Azure Proxy Route
* @description Call any azure endpoint using the proxy.
*
* Just use `{PROXY_BASE_URL}/azure/{endpoint:path}`
*
* Checks if the deployment id in the url is a litellm model name. If so, it will route using the llm_router.allm_passthrough_route.
*/
patch: operations["azure_proxy_route_azure__endpoint__patch"];
trace?: never;
};
"/azure_ai/{endpoint}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Azure Proxy Route
* @description Call any azure endpoint using the proxy.
*
* Just use `{PROXY_BASE_URL}/azure/{endpoint:path}`
*
* Checks if the deployment id in the url is a litellm model name. If so, it will route using the llm_router.allm_passthrough_route.
*/
get: operations["azure_proxy_route_azure_ai__endpoint__get"];
/**
* Azure Proxy Route
* @description Call any azure endpoint using the proxy.
*
* Just use `{PROXY_BASE_URL}/azure/{endpoint:path}`
*
* Checks if the deployment id in the url is a litellm model name. If so, it will route using the llm_router.allm_passthrough_route.
*/
put: operations["azure_proxy_route_azure_ai__endpoint__put"];
/**
* Azure Proxy Route
* @description Call any azure endpoint using the proxy.
*
* Just use `{PROXY_BASE_URL}/azure/{endpoint:path}`
*
* Checks if the deployment id in the url is a litellm model name. If so, it will route using the llm_router.allm_passthrough_route.
*/
post: operations["azure_proxy_route_azure_ai__endpoint__post"];
/**
* Azure Proxy Route
* @description Call any azure endpoint using the proxy.
*
* Just use `{PROXY_BASE_URL}/azure/{endpoint:path}`
*
* Checks if the deployment id in the url is a litellm model name. If so, it will route using the llm_router.allm_passthrough_route.
*/
delete: operations["azure_proxy_route_azure_ai__endpoint__delete"];
options?: never;
head?: never;
/**
* Azure Proxy Route
* @description Call any azure endpoint using the proxy.
*
* Just use `{PROXY_BASE_URL}/azure/{endpoint:path}`
*
* Checks if the deployment id in the url is a litellm model name. If so, it will route using the llm_router.allm_passthrough_route.
*/
patch: operations["azure_proxy_route_azure_ai__endpoint__patch"];
trace?: never;
};
"/azure_speech/{endpoint}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Azure Speech Proxy Route
* @description Pass-through for the Azure AI Speech REST APIs (speech to text), e.g.
* `POST /azure_speech/speech/recognition/conversation/cognitiveservices/v1?language=en-US`
* with the raw audio as the body, or `POST /azure_speech/speechtotext/v3.2/transcriptions`.
*
* The body is forwarded byte for byte and the proxy injects its own
* `Ocp-Apim-Subscription-Key`; the caller's `Authorization` header is the LiteLLM key
* and is never forwarded.
*
* [Docs](https://docs.litellm.ai/docs/pass_through/azure_speech)
*/
get: operations["azure_speech_proxy_route_azure_speech__endpoint__get"];
/**
* Azure Speech Proxy Route
* @description Pass-through for the Azure AI Speech REST APIs (speech to text), e.g.
* `POST /azure_speech/speech/recognition/conversation/cognitiveservices/v1?language=en-US`
* with the raw audio as the body, or `POST /azure_speech/speechtotext/v3.2/transcriptions`.
*
* The body is forwarded byte for byte and the proxy injects its own
* `Ocp-Apim-Subscription-Key`; the caller's `Authorization` header is the LiteLLM key
* and is never forwarded.
*
* [Docs](https://docs.litellm.ai/docs/pass_through/azure_speech)
*/
put: operations["azure_speech_proxy_route_azure_speech__endpoint__put"];
/**
* Azure Speech Proxy Route
* @description Pass-through for the Azure AI Speech REST APIs (speech to text), e.g.
* `POST /azure_speech/speech/recognition/conversation/cognitiveservices/v1?language=en-US`
* with the raw audio as the body, or `POST /azure_speech/speechtotext/v3.2/transcriptions`.
*
* The body is forwarded byte for byte and the proxy injects its own
* `Ocp-Apim-Subscription-Key`; the caller's `Authorization` header is the LiteLLM key
* and is never forwarded.
*
* [Docs](https://docs.litellm.ai/docs/pass_through/azure_speech)
*/
post: operations["azure_speech_proxy_route_azure_speech__endpoint__post"];
/**
* Azure Speech Proxy Route
* @description Pass-through for the Azure AI Speech REST APIs (speech to text), e.g.
* `POST /azure_speech/speech/recognition/conversation/cognitiveservices/v1?language=en-US`
* with the raw audio as the body, or `POST /azure_speech/speechtotext/v3.2/transcriptions`.
*
* The body is forwarded byte for byte and the proxy injects its own
* `Ocp-Apim-Subscription-Key`; the caller's `Authorization` header is the LiteLLM key
* and is never forwarded.
*
* [Docs](https://docs.litellm.ai/docs/pass_through/azure_speech)
*/
delete: operations["azure_speech_proxy_route_azure_speech__endpoint__delete"];
options?: never;
head?: never;
/**
* Azure Speech Proxy Route
* @description Pass-through for the Azure AI Speech REST APIs (speech to text), e.g.
* `POST /azure_speech/speech/recognition/conversation/cognitiveservices/v1?language=en-US`
* with the raw audio as the body, or `POST /azure_speech/speechtotext/v3.2/transcriptions`.
*
* The body is forwarded byte for byte and the proxy injects its own
* `Ocp-Apim-Subscription-Key`; the caller's `Authorization` header is the LiteLLM key
* and is never forwarded.
*
* [Docs](https://docs.litellm.ai/docs/pass_through/azure_speech)
*/
patch: operations["azure_speech_proxy_route_azure_speech__endpoint__patch"];
trace?: never;
};
"/batches": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List Batches
* @description Lists
* This is the equivalent of GET https://api.openai.com/v1/batches/
* Supports Identical Params as: https://platform.openai.com/docs/api-reference/batch/list
*
* Example Curl
* ```
* curl http://localhost:4000/v1/batches?limit=2 -H "Authorization: Bearer $LITELLM_MASTER_KEY" -H "Content-Type: application/json"
* ```
*/
get: operations["list_batches_batches_get"];
put?: never;
/**
* Create Batch
* @description Create large batches of API requests for asynchronous processing.
* This is the equivalent of POST https://api.openai.com/v1/batch
* Supports Identical Params as: https://platform.openai.com/docs/api-reference/batch
*
* Example Curl
* ```
* curl http://localhost:4000/v1/batches -H "Authorization: Bearer $LITELLM_MASTER_KEY" -H "Content-Type: application/json" -d '{
* "input_file_id": "file-abc123",
* "endpoint": "/v1/chat/completions",
* "completion_window": "24h"
* }'
* ```
*/
post: operations["create_batch_batches_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/batches/{batch_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Retrieve Batch
* @description Retrieves a batch.
* This is the equivalent of GET https://api.openai.com/v1/batches/{batch_id}
* Supports Identical Params as: https://platform.openai.com/docs/api-reference/batch/retrieve
*
* Example Curl
* ```
* curl http://localhost:4000/v1/batches/batch_abc123 -H "Authorization: Bearer $LITELLM_MASTER_KEY" -H "Content-Type: application/json"
* ```
*/
get: operations["retrieve_batch_batches__batch_id__get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/batches/{batch_id}/cancel": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Cancel Batch
* @description Cancel a batch.
* This is the equivalent of POST https://api.openai.com/v1/batches/{batch_id}/cancel
*
* Supports Identical Params as: https://platform.openai.com/docs/api-reference/batch/cancel
*
* Example Curl
* ```
* curl http://localhost:4000/v1/batches/batch_abc123/cancel -H "Authorization: Bearer $LITELLM_MASTER_KEY" -H "Content-Type: application/json" -X POST
*
* ```
*/
post: operations["cancel_batch_batches__batch_id__cancel_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/bedrock/{endpoint}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Bedrock Proxy Route
* @description This is the v1 passthrough for Bedrock.
* V2 is handled by the `/bedrock/v2` endpoint.
* [Docs](https://docs.litellm.ai/docs/pass_through/bedrock)
*/
get: operations["bedrock_proxy_route_bedrock__endpoint__get"];
/**
* Bedrock Proxy Route
* @description This is the v1 passthrough for Bedrock.
* V2 is handled by the `/bedrock/v2` endpoint.
* [Docs](https://docs.litellm.ai/docs/pass_through/bedrock)
*/
put: operations["bedrock_proxy_route_bedrock__endpoint__put"];
/**
* Bedrock Proxy Route
* @description This is the v1 passthrough for Bedrock.
* V2 is handled by the `/bedrock/v2` endpoint.
* [Docs](https://docs.litellm.ai/docs/pass_through/bedrock)
*/
post: operations["bedrock_proxy_route_bedrock__endpoint__post"];
/**
* Bedrock Proxy Route
* @description This is the v1 passthrough for Bedrock.
* V2 is handled by the `/bedrock/v2` endpoint.
* [Docs](https://docs.litellm.ai/docs/pass_through/bedrock)
*/
delete: operations["bedrock_proxy_route_bedrock__endpoint__delete"];
options?: never;
head?: never;
/**
* Bedrock Proxy Route
* @description This is the v1 passthrough for Bedrock.
* V2 is handled by the `/bedrock/v2` endpoint.
* [Docs](https://docs.litellm.ai/docs/pass_through/bedrock)
*/
patch: operations["bedrock_proxy_route_bedrock__endpoint__patch"];
trace?: never;
};
"/bespoke/v1/systemone": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/** Bespoke Proxy Route */
post: operations["bespoke_proxy_route_bespoke_v1_systemone_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/budget/delete": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Delete Budget
* @description Delete budget
*
* Parameters:
* - id: str - The budget id to delete
*/
post: operations["delete_budget_budget_delete_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/budget/info": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Info Budget
* @description Get the budget id specific information
*
* Parameters:
* - budgets: List[str] - The list of budget ids to get information for
*/
post: operations["info_budget_budget_info_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/budget/list": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List Budget
* @description List all the created budgets in proxy db. Used on Admin UI.
*/
get: operations["list_budget_budget_list_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/budget/new": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* New Budget
* @description Create a new budget object. Can apply this to teams, orgs, end-users, keys.
*
* Parameters:
* - budget_duration: Optional[str] - Budget reset period ("30d", "1h", etc.)
* - budget_id: Optional[str] - The id of the budget. If not provided, a new id will be generated.
* - max_budget: Optional[float] - The max budget for the budget.
* - soft_budget: Optional[float] - The soft budget for the budget.
* - max_parallel_requests: Optional[int] - The max number of parallel requests for the budget.
* - tpm_limit: Optional[int] - The tokens per minute limit for the budget.
* - rpm_limit: Optional[int] - The requests per minute limit for the budget.
* - tpd_limit: Optional[int] - The tokens per day limit for the budget. Charged by batch submissions instead of tpm_limit/rpm_limit.
* - model_max_budget: Optional[dict] - Specify max budget for a given model. Example: {"openai/gpt-4o-mini": {"max_budget": 100.0, "budget_duration": "1d", "tpm_limit": 100000, "rpm_limit": 100000}}
* - budget_reset_at: Optional[datetime] - Datetime when the initial budget is reset. Default is now.
*/
post: operations["new_budget_budget_new_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/budget/settings": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Budget Settings
* @description Get list of configurable params + current value for a budget item + description of each field
*
* Used on Admin UI.
*
* Query Parameters:
* - budget_id: str - The budget id to get information for
*/
get: operations["budget_settings_budget_settings_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/budget/update": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Update Budget
* @description Update an existing budget object.
*
* Parameters:
* - budget_duration: Optional[str] - Budget reset period ("30d", "1h", etc.)
* - budget_id: Optional[str] - The id of the budget. If not provided, a new id will be generated.
* - max_budget: Optional[float] - The max budget for the budget.
* - soft_budget: Optional[float] - The soft budget for the budget.
* - max_parallel_requests: Optional[int] - The max number of parallel requests for the budget.
* - tpm_limit: Optional[int] - The tokens per minute limit for the budget.
* - rpm_limit: Optional[int] - The requests per minute limit for the budget.
* - tpd_limit: Optional[int] - The tokens per day limit for the budget. Charged by batch submissions instead of tpm_limit/rpm_limit.
* - model_max_budget: Optional[dict] - Specify max budget for a given model. Example: {"openai/gpt-4o-mini": {"max_budget": 100.0, "budget_duration": "1d", "tpm_limit": 100000, "rpm_limit": 100000}}
* - budget_reset_at: Optional[datetime] - Update the Datetime when the budget was last reset.
*/
post: operations["update_budget_budget_update_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/cache/delete": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Cache Delete
* @description Endpoint for deleting a key from the cache. All responses from litellm proxy have `x-litellm-cache-key` in the headers
*
* Parameters:
* - **keys**: *Optional[List[str]]* - A list of keys to delete from the cache. Example {"keys": ["key1", "key2"]}
*
* ```shell
* curl -X POST "http://0.0.0.0:4000/cache/delete" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -d '{"keys": ["key1", "key2"]}'
* ```
*/
post: operations["cache_delete_cache_delete_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/cache/flushall": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Cache Flushall
* @description A function to flush all items from the cache. (All items will be deleted from the cache with this)
* Raises HTTPException if the cache is not initialized or if the cache type does not support flushing.
* Returns a dictionary with the status of the operation.
*
* Usage:
* ```
* curl -X POST http://0.0.0.0:4000/cache/flushall -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*/
post: operations["cache_flushall_cache_flushall_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/cache/ping": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Cache Ping
* @description Endpoint for checking if cache can be pinged
*/
get: operations["cache_ping_cache_ping_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/cache/redis/info": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Cache Redis Info
* @description Endpoint for getting /redis/info
*/
get: operations["cache_redis_info_cache_redis_info_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/cache/settings": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Cache Settings
* @description Get cache configuration and available settings.
*
* Returns:
* - fields: List of all configurable cache settings with their metadata (type, description, default, options)
* - current_values: Current values of cache settings from database
*/
get: operations["get_cache_settings_cache_settings_get"];
put?: never;
/**
* Update Cache Settings
* @description Save cache settings to database and initialize cache.
*
* This endpoint:
* 1. Encrypts sensitive fields (passwords, etc.)
* 2. Saves to LiteLLM_CacheConfig table
* 3. Reinitializes cache with new settings
*/
post: operations["update_cache_settings_cache_settings_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/cache/settings/test": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Test Cache Connection
* @description Test cache connection with provided credentials.
*
* Creates a temporary cache instance and uses its test_connection method
* to verify the credentials work without affecting global state.
*/
post: operations["test_cache_connection_cache_settings_test_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/callback": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Callback
* @description OAuth 2.0 authorization response handler for MCP loopback clients.
*
* Accepts either:
*
* - A successful authorization response (``code`` + ``state``), which is
* forwarded back to the validated client ``redirect_uri`` with the
* original (un-wrapped) ``state``, once the RFC 9207 ``iss`` (when the
* authorization server sent one) matches the issuer /authorize sealed
* into the state.
* - An error response (``error``[+``error_description``/``error_uri``]), per
* RFC 6749 §4.1.2.1. When ``state`` is present and decodes to a trusted
* ``redirect_uri``, the error params are propagated back to the client so
* its OAuth library can surface them. Otherwise we render an HTML error
* page so the user is not left on an opaque 422 / blank screen.
*/
get: operations["callback_callback_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/callbacks/configs": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Callback Configs
* @description Get Available Callback Configurations
*
* Returns the configuration details for all available logging callbacks,
* including supported parameters, field types, and descriptions.
*/
get: operations["get_callback_configs_callbacks_configs_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/callbacks/list": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List Callbacks
* @description View List of Active Logging Callbacks
*/
get: operations["list_callbacks_callbacks_list_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/chat/completions": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Chat Completion
* @description Follows the exact same API spec as `OpenAI's Chat API https://platform.openai.com/docs/api-reference/chat`
*
* ```bash
* curl -X POST http://localhost:4000/v1/chat/completions
* -H "Content-Type: application/json"
* -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* -d '{
* "model": "gpt-4o",
* "messages": [
* {
* "role": "user",
* "content": "Hello!"
* }
* ]
* }'
* ```
*/
post: operations["chat_completion_chat_completions_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/claude-code/marketplace.json": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Marketplace
* @description Serve marketplace.json for Claude Code plugin discovery.
*
* This endpoint is accessed by Claude Code CLI when users run:
* - claude plugin marketplace add <url>
* - claude plugin install <name>@<marketplace>
*
* Without `key` the catalog holds the enabled (public) plugins. With `?key=sk-...`
* the key is authenticated and the catalog also holds the disabled plugins granted
* to it through `object_permission.skills` or an access group on the key or its team.
*
* Returns:
* Marketplace catalog with list of available plugins and their git sources.
*
* Example:
* ```bash
* claude plugin marketplace add http://localhost:4000/claude-code/marketplace.json
* claude plugin marketplace add "http://localhost:4000/claude-code/marketplace.json?key=sk-..."
* claude plugin install my-plugin@litellm
* ```
*/
get: operations["get_marketplace_claude_code_marketplace_json_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/claude-code/plugins": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List Plugins
* @description List all plugins in the marketplace.
*
* Parameters:
* - enabled_only: If true, only return enabled plugins
*
* Returns:
* List of plugins with their metadata.
*/
get: operations["list_plugins_claude_code_plugins_get"];
put?: never;
/**
* Register Plugin
* @description Register a new plugin in the LiteLLM marketplace.
*
* LiteLLM acts as a registry/discovery layer. Plugins are hosted on
* GitHub/GitLab/Bitbucket or as a zip archive on any https host (e.g. S3).
* Claude Code clones the git source or downloads the archive when users install.
*
* This endpoint is create-only and never overwrites. If a plugin with
* the same name already exists it returns 409 Conflict; use
* PUT /claude-code/plugins/{plugin_name} to update an existing plugin.
*
* Requires a proxy admin API key.
*
* Parameters:
* - name: Plugin name (kebab-case)
* - source: Plugin source reference (github, url, git-subdir, or archive format)
* - version: Semantic version (optional)
* - description: Plugin description (optional)
* - author: Author information (optional)
* - homepage: Plugin homepage URL (optional)
* - keywords: Search keywords (optional)
* - category: Plugin category (optional)
* - installation_preference: 'available', 'auto_install', or 'required' (optional)
*
* Returns:
* Registration status (action is always "created") and plugin information.
*
* Example:
* ```bash
* curl -X POST http://localhost:4000/claude-code/plugins \
* -H "Authorization: Bearer sk-..." \
* -H "Content-Type: application/json" \
* -d '{
* "name": "my-plugin",
* "source": {"source": "github", "repo": "org/my-plugin"},
* "version": "1.0.0",
* "description": "My awesome plugin"
* }'
* ```
*/
post: operations["register_plugin_claude_code_plugins_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/claude-code/plugins/{plugin_name}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Plugin
* @description Get details of a specific plugin.
*
* Parameters:
* - plugin_name: The name of the plugin
*
* Returns:
* Plugin details including source and metadata.
*/
get: operations["get_plugin_claude_code_plugins__plugin_name__get"];
/**
* Update Plugin
* @description Update an existing plugin in the LiteLLM marketplace.
*
* The plugin is identified by its name in the path, which is the resource
* identity and cannot be changed here. This is a full replace, not a merge:
* the manifest is rebuilt from the request body, so any optional field left
* out is reset to its default (e.g. an omitted version is cleared, not kept).
* Send the full desired state.
*
* Returns 404 if no plugin with the given name exists; use
* POST /claude-code/plugins to create a new plugin.
*
* Requires a proxy admin API key.
*
* Parameters:
* - plugin_name: Name of the plugin to update (path parameter)
* - source: Plugin source reference (github, url, git-subdir, or archive format)
* - version: Semantic version (optional)
* - description: Plugin description (optional)
* - author: Author information (optional)
* - homepage: Plugin homepage URL (optional)
* - keywords: Search keywords (optional)
* - category: Plugin category (optional)
* - installation_preference: 'available', 'auto_install', or 'required' (optional)
*
* Returns:
* Update status (action is always "updated") and plugin information.
*
* Example:
* ```bash
* curl -X PUT http://localhost:4000/claude-code/plugins/my-plugin \
* -H "Authorization: Bearer sk-..." \
* -H "Content-Type: application/json" \
* -d '{
* "source": {"source": "github", "repo": "org/my-plugin"},
* "version": "2.0.0",
* "description": "My awesome plugin"
* }'
* ```
*/
put: operations["update_plugin_claude_code_plugins__plugin_name__put"];
post?: never;
/**
* Delete Plugin
* @description Delete a plugin from the marketplace.
*
* Requires a proxy admin API key.
*
* Parameters:
* - plugin_name: The name of the plugin to delete
*/
delete: operations["delete_plugin_claude_code_plugins__plugin_name__delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/claude-code/plugins/{plugin_name}/disable": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Disable Plugin
* @description Disable a plugin without deleting it.
*
* Requires a proxy admin API key.
*
* Parameters:
* - plugin_name: The name of the plugin to disable
*/
post: operations["disable_plugin_claude_code_plugins__plugin_name__disable_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/claude-code/plugins/{plugin_name}/enable": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Enable Plugin
* @description Enable a disabled plugin.
*
* Requires a proxy admin API key.
*
* Parameters:
* - plugin_name: The name of the plugin to enable
*/
post: operations["enable_plugin_claude_code_plugins__plugin_name__enable_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/cloudzero/delete": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
post?: never;
/**
* Delete Cloudzero Settings
* @description Delete CloudZero settings from the database.
*
* This endpoint removes the CloudZero configuration (API key, connection ID, timezone)
* from the proxy database. Only the CloudZero settings entry will be deleted;
* other configuration values in the database will remain unchanged.
*
* Only admin users can delete CloudZero settings.
*/
delete: operations["delete_cloudzero_settings_cloudzero_delete_delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/cloudzero/dry-run": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Cloudzero Dry Run Export
* @description Perform a dry run export using the CloudZero logger.
*
* This endpoint uses the CloudZero logger to perform a dry run export,
* which returns the data that would be exported without actually sending it to CloudZero.
*
* Parameters:
* - limit: Optional limit on number of records to process (default: 10000)
*
* Returns:
* - usage_data: Sample of the raw usage data (first 50 records)
* - cbf_data: CloudZero CBF formatted data ready for export
* - summary: Statistics including total cost, tokens, and record counts
*
* Only admin users can perform CloudZero exports.
*/
post: operations["cloudzero_dry_run_export_cloudzero_dry_run_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/cloudzero/export": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Cloudzero Export
* @description Perform an actual export using the CloudZero logger.
*
* This endpoint uses the CloudZero logger to export usage data to CloudZero AnyCost API.
*
* Parameters:
* - limit: Optional limit on number of records to export
* - operation: CloudZero operation type ("replace_hourly" or "sum", default: "replace_hourly")
*
* Only admin users can perform CloudZero exports.
*/
post: operations["cloudzero_export_cloudzero_export_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/cloudzero/init": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Init Cloudzero Settings
* @description Initialize CloudZero settings and store in the database.
*
* This endpoint stores the CloudZero API key, connection ID, and timezone configuration
* in the proxy database for use by the CloudZero logger.
*
* Parameters:
* - api_key: CloudZero API key for authentication
* - connection_id: CloudZero connection ID for data submission
* - timezone: Timezone for date handling (default: UTC)
*
* Only admin users can configure CloudZero settings.
*/
post: operations["init_cloudzero_settings_cloudzero_init_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/cloudzero/settings": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Cloudzero Settings
* @description View current CloudZero settings.
*
* Returns the current CloudZero configuration with the API key masked for security.
* Only the first 4 and last 4 characters of the API key are shown.
* Returns null/empty values when settings are not configured (consistent with other settings endpoints).
*
* Only admin users (Proxy Admin or Admin Viewer) can view CloudZero settings.
*/
get: operations["get_cloudzero_settings_cloudzero_settings_get"];
/**
* Update Cloudzero Settings
* @description Update existing CloudZero settings.
*
* Allows updating individual CloudZero configuration fields without requiring all fields.
* Only provided fields will be updated; others will remain unchanged.
*
* Parameters:
* - api_key: (Optional) New CloudZero API key for authentication
* - connection_id: (Optional) New CloudZero connection ID for data submission
* - timezone: (Optional) New timezone for date handling
*
* Only admin users can update CloudZero settings.
*/
put: operations["update_cloudzero_settings_cloudzero_settings_put"];
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/cohere/{endpoint}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Cohere Proxy Route
* @description [Docs](https://docs.litellm.ai/docs/pass_through/cohere)
*/
get: operations["cohere_proxy_route_cohere__endpoint__get"];
/**
* Cohere Proxy Route
* @description [Docs](https://docs.litellm.ai/docs/pass_through/cohere)
*/
put: operations["cohere_proxy_route_cohere__endpoint__put"];
/**
* Cohere Proxy Route
* @description [Docs](https://docs.litellm.ai/docs/pass_through/cohere)
*/
post: operations["cohere_proxy_route_cohere__endpoint__post"];
/**
* Cohere Proxy Route
* @description [Docs](https://docs.litellm.ai/docs/pass_through/cohere)
*/
delete: operations["cohere_proxy_route_cohere__endpoint__delete"];
options?: never;
head?: never;
/**
* Cohere Proxy Route
* @description [Docs](https://docs.litellm.ai/docs/pass_through/cohere)
*/
patch: operations["cohere_proxy_route_cohere__endpoint__patch"];
trace?: never;
};
"/completions": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Completion
* @description Follows the exact same API spec as `OpenAI's Completions API https://platform.openai.com/docs/api-reference/completions`
*
* ```bash
* curl -X POST http://localhost:4000/v1/completions
* -H "Content-Type: application/json"
* -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* -d '{
* "model": "gpt-3.5-turbo-instruct",
* "prompt": "Once upon a time",
* "max_tokens": 50,
* "temperature": 0.7
* }'
* ```
*/
post: operations["completion_completions_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/compliance/eu-ai-act": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Check Eu Ai Act Compliance
* @description Check EU AI Act compliance for a spend log entry.
*
* Checks:
* - Art. 9: Guardrails applied (any guardrail)
* - Art. 5: Content screened before LLM (pre-call guardrails)
* - Art. 12: Audit record complete (user_id, model, timestamp, guardrail_results)
*/
post: operations["check_eu_ai_act_compliance_compliance_eu_ai_act_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/compliance/gdpr": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Check Gdpr Compliance
* @description Check GDPR compliance for a spend log entry.
*
* Checks:
* - Art. 32: Data protection applied (pre-call guardrails)
* - Art. 5(1)(c): Sensitive data protected (masked/blocked or no issues)
* - Art. 30: Audit record complete (user_id, model, timestamp, guardrail_results)
*/
post: operations["check_gdpr_compliance_compliance_gdpr_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/comprehendmedical": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Comprehend Medical Sdk Proxy Route
* @description AWS-SDK-shaped pass-through for Amazon Comprehend Medical: point the SDK's
* `endpoint_url` at `/comprehendmedical` and the operation is read from the
* `X-Amz-Target` header, per the AWS JSON 1.1 protocol.
*
* [Docs](https://docs.litellm.ai/docs/pass_through/comprehend_medical)
*/
post: operations["comprehend_medical_sdk_proxy_route_comprehendmedical_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/comprehendmedical/{operation}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Comprehend Medical Proxy Route
* @description Pass-through for Amazon Comprehend Medical, e.g. `POST /comprehendmedical/DetectEntitiesV2`.
*
* The request body is forwarded as-is to the AWS JSON 1.1 API and signed with SigV4
* using the proxy's AWS credentials.
*
* [Docs](https://docs.litellm.ai/docs/pass_through/comprehend_medical)
*/
post: operations["comprehend_medical_proxy_route_comprehendmedical__operation__post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/config/block_requests_for_models_without_pricing": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get Block Requests For Models Without Pricing */
get: operations["get_block_requests_for_models_without_pricing_config_block_requests_for_models_without_pricing_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
/** Update Block Requests For Models Without Pricing */
patch: operations["update_block_requests_for_models_without_pricing_config_block_requests_for_models_without_pricing_patch"];
trace?: never;
};
"/config/callback/delete": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Delete Callback
* @description Delete specific logging callback from configuration.
*/
post: operations["delete_callback_config_callback_delete_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/config/cost_discount_config": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Cost Discount Config
* @description Get current cost discount configuration.
*
* Returns the cost_discount_config from litellm_settings.
*/
get: operations["get_cost_discount_config_config_cost_discount_config_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
/**
* Update Cost Discount Config
* @description Update cost discount configuration.
*
* Updates the cost_discount_config in litellm_settings.
* Discounts should be between 0 and 1 (e.g., 0.05 = 5% discount).
* A key may also be <provider>/<model-pattern>, where the pattern is an
* fnmatch glob matched against the model name.
*
* Example:
* ```json
* {
* "vertex_ai": 0.05,
* "vertex_ai/claude-*": 0.20,
* "gemini": 0.05,
* "openai": 0.01
* }
* ```
*/
patch: operations["update_cost_discount_config_config_cost_discount_config_patch"];
trace?: never;
};
"/config/cost_margin_config": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Cost Margin Config
* @description Get current cost margin configuration.
*
* Returns the cost_margin_config from litellm_settings.
*/
get: operations["get_cost_margin_config_config_cost_margin_config_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
/**
* Update Cost Margin Config
* @description Update cost margin configuration.
*
* Updates the cost_margin_config in litellm_settings.
* Margins can be:
* - Percentage: {"openai": 0.10} = 10% margin
* - Fixed amount: {"openai": {"fixed_amount": 0.001}} = $0.001 per request
* - Combined: {"vertex_ai": {"percentage": 0.08, "fixed_amount": 0.0005}}
* - Global: {"global": 0.05} = 5% global margin on all providers
*
* Example:
* ```json
* {
* "global": 0.05,
* "openai": 0.10,
* "anthropic": {"fixed_amount": 0.001},
* "vertex_ai": {"percentage": 0.08, "fixed_amount": 0.0005}
* }
* ```
*/
patch: operations["update_cost_margin_config_config_cost_margin_config_patch"];
trace?: never;
};
"/config/field/delete": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Delete Config General Settings
* @description Delete the db value of this field in litellm general settings. Resets it to it's initial default value on litellm.
*/
post: operations["delete_config_general_settings_config_field_delete_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/config/field/info": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get Config General Settings */
get: operations["get_config_general_settings_config_field_info_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/config/field/update": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Update Config General Settings
* @description Update a specific field in litellm general settings
*/
post: operations["update_config_general_settings_config_field_update_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/config/list": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Config List
* @description List the available fields + current values for a given type of setting (currently just 'general_settings'user_api_key_dict: UserAPIKeyAuth = Depends(user_api_key_auth),)
*/
get: operations["get_config_list_config_list_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/config/pass_through_endpoint": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Pass Through Endpoints
* @description GET configured pass through endpoint.
*
* If no endpoint_id given, return all configured endpoints.
*/
get: operations["get_pass_through_endpoints_config_pass_through_endpoint_get"];
put?: never;
/**
* Create Pass Through Endpoints
* @description Create new pass-through endpoint
*/
post: operations["create_pass_through_endpoints_config_pass_through_endpoint_post"];
/**
* Delete Pass Through Endpoints
* @description Delete a pass-through endpoint by ID.
*
* Returns - the deleted endpoint
*/
delete: operations["delete_pass_through_endpoints_config_pass_through_endpoint_delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/config/pass_through_endpoint/team/{team_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Pass Through Endpoints
* @description GET configured pass through endpoint.
*
* If no endpoint_id given, return all configured endpoints.
*/
get: operations["get_pass_through_endpoints_config_pass_through_endpoint_team__team_id__get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/config/pass_through_endpoint/{endpoint_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Update Pass Through Endpoints
* @description Update a pass-through endpoint by ID.
*/
post: operations["update_pass_through_endpoints_config_pass_through_endpoint__endpoint_id__post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/config/update": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Update Config
* @description For Admin UI - allows admin to update config via UI.
*
* Writes only the sections present in the request body to LiteLLM_Config rows
* (one row per top-level section). Sections the caller did not send are left
* untouched — this endpoint never persists pre-existing YAML values to DB as
* a side effect of an unrelated update.
*/
post: operations["update_config_config_update_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/config/yaml": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Config Yaml Endpoint
* @description This is a mock endpoint, to show what you can set in config.yaml details in the Swagger UI.
*
* Parameters:
*
* The config.yaml object has the following attributes:
* - **model_list**: *Optional[List[ModelParams]]* - A list of supported models on the server, along with model-specific configurations. ModelParams includes "model_name" (name of the model), "litellm_params" (litellm-specific parameters for the model), and "model_info" (additional info about the model such as id, mode, cost per token, etc).
*
* - **litellm_settings**: *Optional[dict]*: Settings for the litellm module. You can specify multiple properties like "drop_params", "set_verbose", "api_base", "cache".
*
* - **general_settings**: *Optional[ConfigGeneralSettings]*: General settings for the server like "completion_model" (default model for chat completion calls), "use_azure_key_vault" (option to load keys from azure key vault), "master_key" (key required for all calls to proxy), and others.
*
* Please, refer to each class's description for a better understanding of the specific attributes within them.
*
* Note: This is a mock endpoint primarily meant for demonstration purposes, and does not actually provide or change any configurations.
*/
get: operations["config_yaml_endpoint_config_yaml_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/config_overrides/cyberark": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Cyberark Config
* @description Get current CyberArk Conjur configuration.
* Returns decrypted values from DB, or falls back to current env vars.
* Sensitive fields are masked before leaving the server.
*/
get: operations["get_cyberark_config_config_overrides_cyberark_get"];
put?: never;
/**
* Update Cyberark Config
* @description Update CyberArk Conjur secret manager configuration.
* Sets environment variables, encrypts sensitive fields, and stores in DB.
* Reinitializes the secret manager on this pod.
*/
post: operations["update_cyberark_config_config_overrides_cyberark_post"];
/**
* Delete Cyberark Config
* @description Delete CyberArk Conjur configuration. Idempotent.
*/
delete: operations["delete_cyberark_config_config_overrides_cyberark_delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/config_overrides/cyberark/test_connection": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Test Cyberark Connection
* @description Test the connection to the currently configured CyberArk Conjur server.
* Uses the already-initialized secret manager client. Does not modify any state.
*/
post: operations["test_cyberark_connection_config_overrides_cyberark_test_connection_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/config_overrides/hashicorp_vault": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Hashicorp Vault Config
* @description Get current Hashicorp Vault configuration.
* Returns decrypted values from DB, or falls back to current env vars.
*/
get: operations["get_hashicorp_vault_config_config_overrides_hashicorp_vault_get"];
put?: never;
/**
* Update Hashicorp Vault Config
* @description Update Hashicorp Vault secret manager configuration.
* Sets environment variables, encrypts sensitive fields, and stores in DB.
* Reinitializes the secret manager on this pod.
*/
post: operations["update_hashicorp_vault_config_config_overrides_hashicorp_vault_post"];
/**
* Delete Hashicorp Vault Config
* @description Delete Hashicorp Vault configuration. Idempotent.
*/
delete: operations["delete_hashicorp_vault_config_config_overrides_hashicorp_vault_delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/config_overrides/hashicorp_vault/test_connection": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Test Hashicorp Vault Connection
* @description Test the connection to the currently configured Hashicorp Vault.
* Uses the already-initialized secret manager client. Does not modify any state.
*/
post: operations["test_hashicorp_vault_connection_config_overrides_hashicorp_vault_test_connection_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/containers": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List Containers
* @description Container list endpoint for retrieving a list of containers.
*
* Follows the OpenAI Containers API spec:
* https://platform.openai.com/docs/api-reference/containers
*
* Example:
* ```bash
* curl -X GET "http://localhost:4000/v1/containers?limit=20&order=desc" -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*
* Or specify provider via header or query param:
* ```bash
* curl -X GET "http://localhost:4000/v1/containers?custom_llm_provider=azure" -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*/
get: operations["list_containers_containers_get"];
put?: never;
/**
* Create Container
* @description Container creation endpoint for creating new containers.
*
* Follows the OpenAI Containers API spec:
* https://platform.openai.com/docs/api-reference/containers
*
* Example:
* ```bash
* curl -X POST "http://localhost:4000/v1/containers" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -H "Content-Type: application/json" -d '{
* "name": "My Container",
* "expires_after": {
* "anchor": "last_active_at",
* "minutes": 20
* }
* }'
* ```
*
* Or specify provider via header:
* ```bash
* curl -X POST "http://localhost:4000/v1/containers" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -H "custom-llm-provider: azure" -H "Content-Type: application/json" -d '{
* "name": "My Container"
* }'
* ```
*/
post: operations["create_container_containers_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/containers/{container_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Retrieve Container
* @description Container retrieve endpoint for getting details of a specific container.
*
* Follows the OpenAI Containers API spec:
* https://platform.openai.com/docs/api-reference/containers
*
* Example:
* ```bash
* curl -X GET "http://localhost:4000/v1/containers/cntr_123" -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*
* Or specify provider via header:
* ```bash
* curl -X GET "http://localhost:4000/v1/containers/cntr_123" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -H "custom-llm-provider: azure"
* ```
*/
get: operations["retrieve_container_containers__container_id__get"];
put?: never;
post?: never;
/**
* Delete Container
* @description Container delete endpoint for deleting a specific container.
*
* Follows the OpenAI Containers API spec:
* https://platform.openai.com/docs/api-reference/containers
*
* Example:
* ```bash
* curl -X DELETE "http://localhost:4000/v1/containers/cntr_123" -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*
* Or specify provider via header:
* ```bash
* curl -X DELETE "http://localhost:4000/v1/containers/cntr_123" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -H "custom-llm-provider: azure"
* ```
*/
delete: operations["delete_container_containers__container_id__delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/containers/{container_id}/files": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Handler Container Id */
get: operations["handler_container_id_containers__container_id__files_get"];
put?: never;
/** Handler Multipart Upload */
post: operations["handler_multipart_upload_containers__container_id__files_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/containers/{container_id}/files/{file_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Handler Container File */
get: operations["handler_container_file_containers__container_id__files__file_id__get"];
put?: never;
post?: never;
/** Handler Container File */
delete: operations["handler_container_file_containers__container_id__files__file_id__delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/containers/{container_id}/files/{file_id}/content": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Handler Binary Content */
get: operations["handler_binary_content_containers__container_id__files__file_id__content_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/coordination_redis/settings": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Coordination Redis Settings
* @description Get the coordination Redis configuration and available settings.
*
* Returns:
* - values: current coordination Redis settings, with password/sentinel_password/url redacted
* - fields: all configurable settings with their metadata (type, description, default, section)
* - source: "coordination_redis" | "cache_backend" | "environment" | null
*/
get: operations["get_coordination_redis_settings_coordination_redis_settings_get"];
put?: never;
/**
* Update Coordination Redis Settings
* @description Save coordination Redis settings under `general_settings.coordination_redis`.
*
* Parameters:
* - settings: dict - Redis connection params (host, port, username, password, url, ssl, startup_nodes, sentinel_nodes, sentinel_password, service_name). Values may be `os.environ/VAR` references, which are stored as written and resolved at startup
*
* The settings are written to the `general_settings` row of LiteLLM_Config,
* which startup merges over the yaml config; the proxy picks them up on its
* next restart.
*/
post: operations["update_coordination_redis_settings_coordination_redis_settings_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/coordination_redis/settings/test": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Check Coordination Redis Connection
* @description Test a coordination Redis connection with the provided credentials.
*
* Parameters:
* - settings: dict - Redis connection params to test. Credential fields sent back as `***REDACTED***` fall back to the saved value
*
* Builds a throwaway client (never touching global state) and pings it.
*/
post: operations["check_coordination_redis_connection_coordination_redis_settings_test_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/cost/estimate": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Estimate Cost
* @description Estimate cost for a given model and token counts.
*
* This endpoint uses the same cost calculation logic as actual requests,
* including any configured margins and discounts.
*
* Parameters:
* - model: Model name (e.g., "gpt-4", "claude-3-opus")
* - input_tokens: Expected input tokens per request
* - output_tokens: Expected output tokens per request
* - cache_read_input_tokens: Cache-read tokens per request, counted within input_tokens (optional)
* - cache_creation_input_tokens: Cache-write tokens per request, counted within input_tokens (optional)
* - reasoning_tokens: Reasoning tokens per request, counted within output_tokens (optional)
* - num_requests_per_day: Number of requests per day (optional)
* - num_requests_per_month: Number of requests per month (optional)
*
* Returns cost breakdown including:
* - Per-request costs (input, output, margin, plus the cache-read, cache-write and reasoning shares)
* - Daily costs (if num_requests_per_day provided)
* - Monthly costs (if num_requests_per_month provided)
*
* Example:
* ```json
* {
* "model": "gpt-4",
* "input_tokens": 1000,
* "cache_read_input_tokens": 800,
* "output_tokens": 500,
* "reasoning_tokens": 200,
* "num_requests_per_day": 100,
* "num_requests_per_month": 3000
* }
* ```
*/
post: operations["estimate_cost_cost_estimate_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/cost/predict-cache": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Predict Cache Cost
* @description Compare the next native Anthropic request on two configured deployment IDs.
*
* Estimates use provider token counting and recent successful cache telemetry for this key.
* Unknown cache state uses the cold scenario when prices/counts are available. Cache observations
* do not guarantee retention. v0 supports one message-content breakpoint, text and client tools;
* system/tool-only breakpoints, thinking, images, nondefault Anthropic versions, beta headers and
* request transforms are unknown.
* Each provider count consumes one RPM unit and holds concurrency capacity; a comparison uses
* up to four counts. The legacy rate limiter returns unknown without contacting the provider.
* This endpoint does not generate tokens, prewarm caches, choose a model or alter routing.
*/
post: operations["predict_cache_cost_cost_predict_cache_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/cost_optimization/prompt_caching/requests": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get Prompt Caching Requests */
get: operations["get_prompt_caching_requests_cost_optimization_prompt_caching_requests_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/credentials": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Credentials
* @description [BETA] endpoint. This might change unexpectedly.
*/
get: operations["get_credentials_credentials_get"];
put?: never;
/**
* Create Credential
* @description [BETA] endpoint. This might change unexpectedly.
* Stores credential in DB.
* Reloads credentials in memory.
*/
post: operations["create_credential_credentials_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/credentials/by_model/{model_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Credential By Model
* @description [BETA] endpoint. This might change unexpectedly.
*/
get: operations["get_credential_by_model_credentials_by_model__model_id__get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/credentials/by_name/{credential_name}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Credential By Name
* @description [BETA] endpoint. This might change unexpectedly.
*/
get: operations["get_credential_by_name_credentials_by_name__credential_name__get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/credentials/migrate-encryption": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Migrate Encryption Endpoint
* @description Re-encrypt all at-rest credentials into the AES-256-GCM (``v2:gcm:``) format.
*
* Admin only. Requires ``general_settings.encryption_algorithm: aes-256-gcm``.
* Idempotent and resumable — re-running skips already-migrated values. Pass
* ``dry_run=true`` for a non-mutating scan (equivalent to ``--check``).
*/
post: operations["migrate_encryption_endpoint_credentials_migrate_encryption_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/credentials/migrate-encryption/check": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Check Encryption Endpoint
* @description Read-only residual scan for compliance attestation. Reports how many at-rest
* values are still in the legacy format. ``residual_legacy == 0`` attests no
* legacy ciphertext remains. Admin only; performs no writes.
*/
get: operations["check_encryption_endpoint_credentials_migrate_encryption_check_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/credentials/user_connections": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List User Connections
* @description List the calling user's per-user provider connections.
*/
get: operations["list_user_connections_credentials_user_connections_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/credentials/{credential_name}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
post?: never;
/**
* Delete Credential
* @description [BETA] endpoint. This might change unexpectedly.
*/
delete: operations["delete_credential_credentials__credential_name__delete"];
options?: never;
head?: never;
/**
* Update Credential
* @description [BETA] endpoint. This might change unexpectedly.
*/
patch: operations["update_credential_credentials__credential_name__patch"];
trace?: never;
};
"/credentials/{credential_name}/jwks": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Credential Internal Issuer Jwks
* @description Export the public JWKS for an anthropic ``internal_issuer`` credential, so the operator can
* register it on the Anthropic federation issuer from the UI. Never touches the private signing
* key: only its derived public JWKS leaves this process. 404s for any other credential shape.
*/
get: operations["get_credential_internal_issuer_jwks_credentials__credential_name__jwks_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/credentials/{credential_name}/user_connection": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
post?: never;
/**
* Delete User Connection
* @description Disconnect the calling user's stored GitHub token for a per-user credential. Idempotent.
*/
delete: operations["delete_user_connection_credentials__credential_name__user_connection_delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/credentials/{credential_name}/user_connection/poll": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Poll User Connection
* @description Poll the device flow once and persist the connection on completion.
*/
post: operations["poll_user_connection_credentials__credential_name__user_connection_poll_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/credentials/{credential_name}/user_connection/start": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Start User Connection
* @description Begin a GitHub device flow for the calling user's connection to a per-user credential.
*/
post: operations["start_user_connection_credentials__credential_name__user_connection_start_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/cursor/chat/completions": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Cursor Chat Completions
* @description Cursor BYOK endpoint. Accepts both request shapes Cursor sends to its OpenAI-compatible
* base URL and always answers in chat completions format.
*
* Cursor agent mode sends Responses API format bodies (`input`, flat tool defs, `reasoning`,
* custom tools) to the chat/completions path while expecting chat completions responses;
* those are routed through the Responses API pipeline and converted back. Genuine chat
* completions bodies (`messages` present) are routed through the standard chat completions
* pipeline, after normalizing each level of the `tools` array and `tool_choice` to the chat
* completions shapes OpenAI requires. Cursor mixes Responses API shapes into chat bodies
* per level, independently: a flat tool def (`{"type": "custom", "name": "ApplyPatch", ...}`)
* gets nested under `custom`, and a flat grammar format
* (`{"type": "grammar", "definition", "syntax"}`) gets wrapped as
* `{"type": "grammar", "grammar": {...}}` wherever it appears, including inside tool defs
* Cursor already sent pre-nested.
*
* ```bash
* curl -X POST http://localhost:4000/cursor/chat/completions -H "Content-Type: application/json" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -d '{
* "model": "gpt-4o",
* "input": [{"role": "user", "content": "Hello"}]
* }'
* Responds back in chat completions format.
* ```
*/
post: operations["cursor_chat_completions_cursor_chat_completions_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/cursor/models": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Cursor Model List
* @description OpenAI-compatible model listing for the Cursor BYOK base URL.
*
* Clients pointed at `<proxy>/cursor` as an OpenAI-compatible base URL resolve and
* verify models via `GET {base}/models` (the OpenAI SDK contract). Without this
* route those requests fall through to the Cursor Cloud Agents passthrough, which
* demands a Cursor API key and 401s, so key verification silently fails before any
* chat request is ever sent. Delegates to the standard `/v1/models` handler.
*/
get: operations["cursor_model_list_cursor_models_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/cursor/v1/models": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Cursor Model List
* @description OpenAI-compatible model listing for the Cursor BYOK base URL.
*
* Clients pointed at `<proxy>/cursor` as an OpenAI-compatible base URL resolve and
* verify models via `GET {base}/models` (the OpenAI SDK contract). Without this
* route those requests fall through to the Cursor Cloud Agents passthrough, which
* demands a Cursor API key and 401s, so key verification silently fails before any
* chat request is ever sent. Delegates to the standard `/v1/models` handler.
*/
get: operations["cursor_model_list_cursor_v1_models_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/cursor/{endpoint}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Cursor Proxy Route
* @description Pass-through endpoint for the Cursor Cloud Agents API.
*
* Supports all Cursor Cloud Agents endpoints:
* - GET /v0/agents — List agents
* - POST /v0/agents — Launch an agent
* - GET /v0/agents/{id} — Agent status
* - GET /v0/agents/{id}/conversation — Agent conversation
* - POST /v0/agents/{id}/followup — Add follow-up
* - POST /v0/agents/{id}/stop — Stop an agent
* - DELETE /v0/agents/{id} — Delete an agent
* - GET /v0/me — API key info
* - GET /v0/models — List models
* - GET /v0/repositories — List GitHub repositories
*
* Uses Basic Authentication (base64-encoded `API_KEY:`).
*
* Credential lookup order:
* 1. passthrough_endpoint_router (config.yaml deployments with use_in_pass_through)
* 2. litellm.credential_list (credentials added via UI)
* 3. CURSOR_API_KEY environment variable
*/
get: operations["cursor_proxy_route_cursor__endpoint__get"];
/**
* Cursor Proxy Route
* @description Pass-through endpoint for the Cursor Cloud Agents API.
*
* Supports all Cursor Cloud Agents endpoints:
* - GET /v0/agents — List agents
* - POST /v0/agents — Launch an agent
* - GET /v0/agents/{id} — Agent status
* - GET /v0/agents/{id}/conversation — Agent conversation
* - POST /v0/agents/{id}/followup — Add follow-up
* - POST /v0/agents/{id}/stop — Stop an agent
* - DELETE /v0/agents/{id} — Delete an agent
* - GET /v0/me — API key info
* - GET /v0/models — List models
* - GET /v0/repositories — List GitHub repositories
*
* Uses Basic Authentication (base64-encoded `API_KEY:`).
*
* Credential lookup order:
* 1. passthrough_endpoint_router (config.yaml deployments with use_in_pass_through)
* 2. litellm.credential_list (credentials added via UI)
* 3. CURSOR_API_KEY environment variable
*/
put: operations["cursor_proxy_route_cursor__endpoint__put"];
/**
* Cursor Proxy Route
* @description Pass-through endpoint for the Cursor Cloud Agents API.
*
* Supports all Cursor Cloud Agents endpoints:
* - GET /v0/agents — List agents
* - POST /v0/agents — Launch an agent
* - GET /v0/agents/{id} — Agent status
* - GET /v0/agents/{id}/conversation — Agent conversation
* - POST /v0/agents/{id}/followup — Add follow-up
* - POST /v0/agents/{id}/stop — Stop an agent
* - DELETE /v0/agents/{id} — Delete an agent
* - GET /v0/me — API key info
* - GET /v0/models — List models
* - GET /v0/repositories — List GitHub repositories
*
* Uses Basic Authentication (base64-encoded `API_KEY:`).
*
* Credential lookup order:
* 1. passthrough_endpoint_router (config.yaml deployments with use_in_pass_through)
* 2. litellm.credential_list (credentials added via UI)
* 3. CURSOR_API_KEY environment variable
*/
post: operations["cursor_proxy_route_cursor__endpoint__post"];
/**
* Cursor Proxy Route
* @description Pass-through endpoint for the Cursor Cloud Agents API.
*
* Supports all Cursor Cloud Agents endpoints:
* - GET /v0/agents — List agents
* - POST /v0/agents — Launch an agent
* - GET /v0/agents/{id} — Agent status
* - GET /v0/agents/{id}/conversation — Agent conversation
* - POST /v0/agents/{id}/followup — Add follow-up
* - POST /v0/agents/{id}/stop — Stop an agent
* - DELETE /v0/agents/{id} — Delete an agent
* - GET /v0/me — API key info
* - GET /v0/models — List models
* - GET /v0/repositories — List GitHub repositories
*
* Uses Basic Authentication (base64-encoded `API_KEY:`).
*
* Credential lookup order:
* 1. passthrough_endpoint_router (config.yaml deployments with use_in_pass_through)
* 2. litellm.credential_list (credentials added via UI)
* 3. CURSOR_API_KEY environment variable
*/
delete: operations["cursor_proxy_route_cursor__endpoint__delete"];
options?: never;
head?: never;
/**
* Cursor Proxy Route
* @description Pass-through endpoint for the Cursor Cloud Agents API.
*
* Supports all Cursor Cloud Agents endpoints:
* - GET /v0/agents — List agents
* - POST /v0/agents — Launch an agent
* - GET /v0/agents/{id} — Agent status
* - GET /v0/agents/{id}/conversation — Agent conversation
* - POST /v0/agents/{id}/followup — Add follow-up
* - POST /v0/agents/{id}/stop — Stop an agent
* - DELETE /v0/agents/{id} — Delete an agent
* - GET /v0/me — API key info
* - GET /v0/models — List models
* - GET /v0/repositories — List GitHub repositories
*
* Uses Basic Authentication (base64-encoded `API_KEY:`).
*
* Credential lookup order:
* 1. passthrough_endpoint_router (config.yaml deployments with use_in_pass_through)
* 2. litellm.credential_list (credentials added via UI)
* 3. CURSOR_API_KEY environment variable
*/
patch: operations["cursor_proxy_route_cursor__endpoint__patch"];
trace?: never;
};
"/customer/block": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Block User
* @description [BETA] Reject calls with this end-user id
*
* Parameters:
* - user_ids (List[str], required): The unique `user_id`s for the users to block
*
* (any /chat/completion call with this user={end-user-id} param, will be rejected.)
*
* ```
* curl -X POST "http://0.0.0.0:8000/user/block"
* -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* -d '{
* "user_ids": [<user_id>, ...]
* }'
* ```
*/
post: operations["block_user_customer_block_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/customer/daily/activity": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Customer Daily Activity
* @description Get daily activity for specific organizations or all accessible organizations.
*/
get: operations["get_customer_daily_activity_customer_daily_activity_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/customer/daily/activity/aggregated": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get Customer Daily Activity Aggregated */
get: operations["get_customer_daily_activity_aggregated_customer_daily_activity_aggregated_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/customer/daily/activity/aggregated/keys": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get Customer Daily Activity Aggregated Keys */
get: operations["get_customer_daily_activity_aggregated_keys_customer_daily_activity_aggregated_keys_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/customer/daily/activity/aggregated/model_top_keys": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get Customer Daily Activity Model Top Keys */
get: operations["get_customer_daily_activity_model_top_keys_customer_daily_activity_aggregated_model_top_keys_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/customer/daily/activity/aggregated/search": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get Customer Daily Activity Aggregated Search */
get: operations["get_customer_daily_activity_aggregated_search_customer_daily_activity_aggregated_search_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/customer/daily/activity/export": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get Customer Daily Activity Export */
get: operations["get_customer_daily_activity_export_customer_daily_activity_export_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/customer/delete": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Delete End User
* @description Delete multiple end-users.
*
* Parameters:
* - user_ids (List[str], required): The unique `user_id`s for the users to delete
*
* Example curl:
* ```
* curl --location 'http://0.0.0.0:4000/customer/delete' --header "Authorization: Bearer $LITELLM_MASTER_KEY" --header 'Content-Type: application/json' --data '{
* "user_ids" :["ishaan-jaff-5"]
* }'
*
* See below for all params
* ```
*/
post: operations["delete_end_user_customer_delete_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/customer/info": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* End User Info
* @description Get information about an end-user. An `end_user` is a customer (external user) of the proxy.
*
* Parameters:
* - end_user_id (str, required): The unique identifier for the end-user
*
* Example curl:
* ```
* curl -X GET 'http://localhost:4000/customer/info?end_user_id=test-litellm-user-4' -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*/
get: operations["end_user_info_customer_info_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/customer/list": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List End User
* @description [Admin-only] List all available customers
*
* Example curl:
* ```
* curl --location --request GET 'http://0.0.0.0:4000/customer/list' --header "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*/
get: operations["list_end_user_customer_list_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/customer/new": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* New End User
* @description Allow creating a new Customer
*
*
* Parameters:
* - user_id: str - The unique identifier for the user.
* - alias: Optional[str] - A human-friendly alias for the user.
* - blocked: bool - Flag to allow or disallow requests for this end-user. Default is False.
* - max_budget: Optional[float] - The maximum budget allocated to the user. Either 'max_budget' or 'budget_id' should be provided, not both.
* - budget_id: Optional[str] - The identifier for an existing budget allocated to the user. Either 'max_budget' or 'budget_id' should be provided, not both.
* - allowed_model_region: Optional[Union[Literal["eu"], Literal["us"]]] - Require all user requests to use models in this specific region.
* - default_model: Optional[str] - If no equivalent model in the allowed region, default all requests to this model.
* - models: Optional[list[str]] - Restrict this customer's access to the listed models.
* - metadata: Optional[dict] = Metadata for customer, store information for customer. Example metadata = {"data_training_opt_out": True}
* - budget_duration: Optional[str] - Budget is reset at the end of specified duration. If not set, budget is never reset. You can set duration as seconds ("30s"), minutes ("30m"), hours ("30h"), days ("30d").
* - tpm_limit: Optional[int] - [Not Implemented Yet] Specify tpm limit for a given customer (Tokens per minute)
* - rpm_limit: Optional[int] - [Not Implemented Yet] Specify rpm limit for a given customer (Requests per minute)
* - tpd_limit: Optional[int] - Specify tpd limit for a given customer (Tokens per day). Batch submissions are charged against it instead of tpm_limit/rpm_limit
* - model_max_budget: Optional[dict] - [Not Implemented Yet] Specify max budget for a given model. Example: {"openai/gpt-4o-mini": {"max_budget": 100.0, "budget_duration": "1d"}}
* - max_parallel_requests: Optional[int] - [Not Implemented Yet] Specify max parallel requests for a given customer.
* - soft_budget: Optional[float] - [Not Implemented Yet] Get alerts when customer crosses given budget, doesn't block requests.
* - spend: Optional[float] - Specify initial spend for a given customer.
* - budget_reset_at: Optional[str] - Specify the date and time when the budget should be reset.
* - object_permission: Optional[LiteLLM_ObjectPermissionBase] - Customer-specific object permissions to control access to resources.
* Supported fields:
* * mcp_servers: List[str] - List of allowed MCP server IDs
* * mcp_access_groups: List[str] - List of MCP access group names
* * mcp_tool_permissions: Dict[str, List[str]] - Map of server ID to allowed tool names (e.g., {"server_1": ["tool_a", "tool_b"]})
* * vector_stores: List[str] - List of allowed vector store IDs
* * agents: List[str] - List of allowed agent IDs
* * agent_access_groups: List[str] - List of agent access group names
* Example: {"mcp_servers": ["server_1", "server_2"], "vector_stores": ["vector_store_1"], "agents": ["agent_1"]}
* IF null or {} then no object-level restrictions apply.
*
*
* - Allow specifying allowed regions
* - Allow specifying default model
*
* Example curl:
* ```
* curl --location 'http://0.0.0.0:4000/customer/new' --header "Authorization: Bearer $LITELLM_MASTER_KEY" --header 'Content-Type: application/json' --data '{
* "user_id" : "ishaan-jaff-3",
* "allowed_region": "eu",
* "budget_id": "free_tier",
* "models": ["gpt-4o-mini"],
* "default_model": "azure/gpt-3.5-turbo-eu"
* }'
*
* # With object permissions
* curl -L -X POST 'http://localhost:4000/customer/new' -H "Authorization: Bearer $LITELLM_MASTER_KEY" -H 'Content-Type: application/json' -d '{
* "user_id": "user_1",
* "object_permission": {
* "mcp_servers": ["server_1"],
* "mcp_access_groups": ["public_group"],
* "vector_stores": ["vector_store_1"]
* }
* }'
*
* # return end-user object
* ```
*
* NOTE: This used to be called `/end_user/new`, we will still be maintaining compatibility for /end_user/XXX for these endpoints
*/
post: operations["new_end_user_customer_new_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/customer/unblock": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Unblock User
* @description [BETA] Unblock calls with this user id
*
* Example
* ```
* curl -X POST "http://0.0.0.0:8000/user/unblock"
* -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* -d '{
* "user_ids": [<user_id>, ...]
* }'
* ```
*/
post: operations["unblock_user_customer_unblock_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/customer/update": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Update End User
* @description Example curl
*
* Parameters:
* - user_id: str
* - alias: Optional[str] = None # human-friendly alias
* - blocked: bool = False # allow/disallow requests for this end-user
* - max_budget: Optional[float] = None
* - budget_id: Optional[str] = None # give either a budget_id or max_budget
* - allowed_model_region: Optional[AllowedModelRegion] = (
* None # require all user requests to use models in this specific region
* )
* - default_model: Optional[str] = (
* None # if no equivalent model in allowed region - default all requests to this model
* )
* - models: Optional[list[str]] = None # omitted or null leaves the allowlist unchanged; an empty list clears it
* - object_permission: Optional[LiteLLM_ObjectPermissionBase] - Customer-specific object permissions to control access to resources.
* Supported fields:
* * mcp_servers: List[str] - List of allowed MCP server IDs
* * mcp_access_groups: List[str] - List of MCP access group names
* * mcp_tool_permissions: Dict[str, List[str]] - Map of server ID to allowed tool names
* * vector_stores: List[str] - List of allowed vector store IDs
* * agents: List[str] - List of allowed agent IDs
* * agent_access_groups: List[str] - List of agent access group names
* Example: {"mcp_servers": ["server_1"], "vector_stores": ["vector_store_1"]}
* IF null or {} then no object-level restrictions apply.
*
* Example curl:
* ```
* curl --location 'http://0.0.0.0:4000/customer/update' --header "Authorization: Bearer $LITELLM_MASTER_KEY" --header 'Content-Type: application/json' --data '{
* "user_id": "test-litellm-user-4",
* "budget_id": "paid_tier",
* "models": ["gpt-4o-mini"]
* }'
*
* # Updating object permissions
* curl -L -X POST 'http://localhost:4000/customer/update' --header "Authorization: Bearer $LITELLM_MASTER_KEY" --header 'Content-Type: application/json' --data '{
* "user_id": "user_1",
* "object_permission": {
* "mcp_servers": ["server_3"],
* "vector_stores": ["vector_store_2", "vector_store_3"]
* }
* }'
*
* See below for all params
* ```
*/
post: operations["update_end_user_customer_update_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/debug/asyncio-tasks": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Active Tasks Stats
* @description Returns:
* total_active_tasks: int
* by_name: { coroutine_name: count }
*/
get: operations["get_active_tasks_stats_debug_asyncio_tasks_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/debug/memory/details": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Memory Details
* @description Get detailed memory diagnostics for deep debugging.
*
* Returns:
* - worker_pid: Process ID
* - process_memory: RAM usage, virtual memory, file handles, threads
* - garbage_collector: GC thresholds, counts, collection history
* - objects: Total tracked objects and top object types
* - uncollectable: Objects that can't be garbage collected (potential leaks)
* - cache_memory: Memory usage of user_api_key, router, and logging caches
* - router_memory: Memory usage of router components (model_list, deployment_names, etc.)
*
* Query Parameters:
* - top_n: Number of top object types to return (default: 20)
* - include_process_info: Include process-level memory info using psutil (default: true)
*
* Example usage:
* curl "http://localhost:4000/debug/memory/details?top_n=30" -H "Authorization: Bearer $LITELLM_MASTER_KEY"
*
* All memory sizes are reported in both bytes and MB.
*/
get: operations["get_memory_details_debug_memory_details_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/debug/memory/gc/configure": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Configure Gc Thresholds Endpoint
* @description Configure Python garbage collection thresholds.
*
* Lower thresholds mean more frequent GC cycles (less memory, more CPU overhead).
* Higher thresholds mean less frequent GC cycles (more memory, less CPU overhead).
*
* Returns:
* - message: Confirmation message
* - previous_thresholds: Old threshold values
* - new_thresholds: New threshold values
* - objects_awaiting_collection: Current object count in gen-0
* - tip: Hint about when next collection will occur
*
* Query Parameters:
* - generation_0: Number of allocations before gen-0 collection (default: 700)
* - generation_1: Number of gen-0 collections before gen-1 collection (default: 10)
* - generation_2: Number of gen-1 collections before gen-2 collection (default: 10)
*
* Example for more aggressive collection:
* curl -X POST "http://localhost:4000/debug/memory/gc/configure?generation_0=500" -H "Authorization: Bearer $LITELLM_MASTER_KEY"
*
* Example for less aggressive collection:
* curl -X POST "http://localhost:4000/debug/memory/gc/configure?generation_0=1000" -H "Authorization: Bearer $LITELLM_MASTER_KEY"
*
* Monitor memory usage with GET /debug/memory/summary after changes.
*/
post: operations["configure_gc_thresholds_endpoint_debug_memory_gc_configure_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/debug/memory/summary": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Memory Summary
* @description Get simplified memory usage summary for the proxy.
*
* Returns:
* - worker_pid: Process ID
* - hostname: Host (the pod on Kubernetes) the worker runs on
* - status: Overall health based on memory usage
* - memory: Process memory usage and RAM info
* - caches: Cache item counts and descriptions
* - garbage_collector: GC status and pending object counts
*
* Example usage:
* curl http://localhost:4000/debug/memory/summary -H "Authorization: Bearer $LITELLM_MASTER_KEY"
*
* For detailed analysis, call GET /debug/memory/details
* For cache management, use the cache management endpoints
*/
get: operations["get_memory_summary_debug_memory_summary_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/debug/report": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Debug Report
* @description The same LiteLLM-owned environment facts the bug report link puts in a GitHub issue:
* versions, deployment kind, and config flags whose keys and values LiteLLM defines.
* Nothing from the operator's config values, request data, or errors
*
* Example usage:
* curl http://localhost:4000/debug/report -H "Authorization: Bearer $LITELLM_MASTER_KEY"
*/
get: operations["get_debug_report_debug_report_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/decisions": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Ask questions about an input (OpenAI Decisions format)
* @description Judge an `input` against 1 to 128 `questions` and get the answers back in the same order.
*
* Question types are `predicate` (yes/no with `probability`), `choice` (one of the `choices`, with
* `probabilities`), `score` (an index into `levels`, with `probabilities`) and `box` (the image region the
* `instructions` describe, with `confidence`). Requests that break these rules are rejected with a 400 before
* any provider is called. `model` is any decision model in the proxy
* model list. System One decision models accept this format too, LiteLLM translates the request and the
* answers. Streaming is not supported.
*
* [Docs](https://docs.litellm.ai/docs/decisions)
*/
post: operations["decisions_decisions_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/delete/allowed_ip": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/** Delete Allowed Ip */
post: operations["delete_allowed_ip_delete_allowed_ip_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/email/event_settings": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Email Event Settings
* @description Get all email event settings
*/
get: operations["get_email_event_settings_email_event_settings_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
/**
* Update Event Settings
* @description Update the settings for email events
*/
patch: operations["update_event_settings_email_event_settings_patch"];
trace?: never;
};
"/email/event_settings/reset": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Reset Event Settings
* @description Reset all email event settings to default (new user invitations on, virtual key creation off)
*/
post: operations["reset_event_settings_email_event_settings_reset_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/embeddings": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Embeddings
* @description Follows the exact same API spec as `OpenAI's Embeddings API https://platform.openai.com/docs/api-reference/embeddings`
*
* ```bash
* curl -X POST http://localhost:4000/v1/embeddings
* -H "Content-Type: application/json"
* -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* -d '{
* "model": "text-embedding-ada-002",
* "input": "The quick brown fox jumps over the lazy dog"
* }'
* ```
*/
post: operations["embeddings_embeddings_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/end_user/block": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Block User
* @description [BETA] Reject calls with this end-user id
*
* Parameters:
* - user_ids (List[str], required): The unique `user_id`s for the users to block
*
* (any /chat/completion call with this user={end-user-id} param, will be rejected.)
*
* ```
* curl -X POST "http://0.0.0.0:8000/user/block"
* -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* -d '{
* "user_ids": [<user_id>, ...]
* }'
* ```
*/
post: operations["block_user_end_user_block_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/end_user/daily/activity": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Customer Daily Activity
* @description Get daily activity for specific organizations or all accessible organizations.
*/
get: operations["get_customer_daily_activity_end_user_daily_activity_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/end_user/daily/activity/aggregated": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get Customer Daily Activity Aggregated */
get: operations["get_customer_daily_activity_aggregated_end_user_daily_activity_aggregated_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/end_user/daily/activity/aggregated/keys": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get Customer Daily Activity Aggregated Keys */
get: operations["get_customer_daily_activity_aggregated_keys_end_user_daily_activity_aggregated_keys_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/end_user/daily/activity/aggregated/model_top_keys": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get Customer Daily Activity Model Top Keys */
get: operations["get_customer_daily_activity_model_top_keys_end_user_daily_activity_aggregated_model_top_keys_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/end_user/daily/activity/aggregated/search": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get Customer Daily Activity Aggregated Search */
get: operations["get_customer_daily_activity_aggregated_search_end_user_daily_activity_aggregated_search_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/end_user/daily/activity/export": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get Customer Daily Activity Export */
get: operations["get_customer_daily_activity_export_end_user_daily_activity_export_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/end_user/delete": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Delete End User
* @description Delete multiple end-users.
*
* Parameters:
* - user_ids (List[str], required): The unique `user_id`s for the users to delete
*
* Example curl:
* ```
* curl --location 'http://0.0.0.0:4000/customer/delete' --header "Authorization: Bearer $LITELLM_MASTER_KEY" --header 'Content-Type: application/json' --data '{
* "user_ids" :["ishaan-jaff-5"]
* }'
*
* See below for all params
* ```
*/
post: operations["delete_end_user_end_user_delete_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/end_user/info": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* End User Info
* @description Get information about an end-user. An `end_user` is a customer (external user) of the proxy.
*
* Parameters:
* - end_user_id (str, required): The unique identifier for the end-user
*
* Example curl:
* ```
* curl -X GET 'http://localhost:4000/customer/info?end_user_id=test-litellm-user-4' -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*/
get: operations["end_user_info_end_user_info_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/end_user/list": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List End User
* @description [Admin-only] List all available customers
*
* Example curl:
* ```
* curl --location --request GET 'http://0.0.0.0:4000/customer/list' --header "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*/
get: operations["list_end_user_end_user_list_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/end_user/new": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* New End User
* @description Allow creating a new Customer
*
*
* Parameters:
* - user_id: str - The unique identifier for the user.
* - alias: Optional[str] - A human-friendly alias for the user.
* - blocked: bool - Flag to allow or disallow requests for this end-user. Default is False.
* - max_budget: Optional[float] - The maximum budget allocated to the user. Either 'max_budget' or 'budget_id' should be provided, not both.
* - budget_id: Optional[str] - The identifier for an existing budget allocated to the user. Either 'max_budget' or 'budget_id' should be provided, not both.
* - allowed_model_region: Optional[Union[Literal["eu"], Literal["us"]]] - Require all user requests to use models in this specific region.
* - default_model: Optional[str] - If no equivalent model in the allowed region, default all requests to this model.
* - models: Optional[list[str]] - Restrict this customer's access to the listed models.
* - metadata: Optional[dict] = Metadata for customer, store information for customer. Example metadata = {"data_training_opt_out": True}
* - budget_duration: Optional[str] - Budget is reset at the end of specified duration. If not set, budget is never reset. You can set duration as seconds ("30s"), minutes ("30m"), hours ("30h"), days ("30d").
* - tpm_limit: Optional[int] - [Not Implemented Yet] Specify tpm limit for a given customer (Tokens per minute)
* - rpm_limit: Optional[int] - [Not Implemented Yet] Specify rpm limit for a given customer (Requests per minute)
* - tpd_limit: Optional[int] - Specify tpd limit for a given customer (Tokens per day). Batch submissions are charged against it instead of tpm_limit/rpm_limit
* - model_max_budget: Optional[dict] - [Not Implemented Yet] Specify max budget for a given model. Example: {"openai/gpt-4o-mini": {"max_budget": 100.0, "budget_duration": "1d"}}
* - max_parallel_requests: Optional[int] - [Not Implemented Yet] Specify max parallel requests for a given customer.
* - soft_budget: Optional[float] - [Not Implemented Yet] Get alerts when customer crosses given budget, doesn't block requests.
* - spend: Optional[float] - Specify initial spend for a given customer.
* - budget_reset_at: Optional[str] - Specify the date and time when the budget should be reset.
* - object_permission: Optional[LiteLLM_ObjectPermissionBase] - Customer-specific object permissions to control access to resources.
* Supported fields:
* * mcp_servers: List[str] - List of allowed MCP server IDs
* * mcp_access_groups: List[str] - List of MCP access group names
* * mcp_tool_permissions: Dict[str, List[str]] - Map of server ID to allowed tool names (e.g., {"server_1": ["tool_a", "tool_b"]})
* * vector_stores: List[str] - List of allowed vector store IDs
* * agents: List[str] - List of allowed agent IDs
* * agent_access_groups: List[str] - List of agent access group names
* Example: {"mcp_servers": ["server_1", "server_2"], "vector_stores": ["vector_store_1"], "agents": ["agent_1"]}
* IF null or {} then no object-level restrictions apply.
*
*
* - Allow specifying allowed regions
* - Allow specifying default model
*
* Example curl:
* ```
* curl --location 'http://0.0.0.0:4000/customer/new' --header "Authorization: Bearer $LITELLM_MASTER_KEY" --header 'Content-Type: application/json' --data '{
* "user_id" : "ishaan-jaff-3",
* "allowed_region": "eu",
* "budget_id": "free_tier",
* "models": ["gpt-4o-mini"],
* "default_model": "azure/gpt-3.5-turbo-eu"
* }'
*
* # With object permissions
* curl -L -X POST 'http://localhost:4000/customer/new' -H "Authorization: Bearer $LITELLM_MASTER_KEY" -H 'Content-Type: application/json' -d '{
* "user_id": "user_1",
* "object_permission": {
* "mcp_servers": ["server_1"],
* "mcp_access_groups": ["public_group"],
* "vector_stores": ["vector_store_1"]
* }
* }'
*
* # return end-user object
* ```
*
* NOTE: This used to be called `/end_user/new`, we will still be maintaining compatibility for /end_user/XXX for these endpoints
*/
post: operations["new_end_user_end_user_new_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/end_user/unblock": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Unblock User
* @description [BETA] Unblock calls with this user id
*
* Example
* ```
* curl -X POST "http://0.0.0.0:8000/user/unblock"
* -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* -d '{
* "user_ids": [<user_id>, ...]
* }'
* ```
*/
post: operations["unblock_user_end_user_unblock_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/end_user/update": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Update End User
* @description Example curl
*
* Parameters:
* - user_id: str
* - alias: Optional[str] = None # human-friendly alias
* - blocked: bool = False # allow/disallow requests for this end-user
* - max_budget: Optional[float] = None
* - budget_id: Optional[str] = None # give either a budget_id or max_budget
* - allowed_model_region: Optional[AllowedModelRegion] = (
* None # require all user requests to use models in this specific region
* )
* - default_model: Optional[str] = (
* None # if no equivalent model in allowed region - default all requests to this model
* )
* - models: Optional[list[str]] = None # omitted or null leaves the allowlist unchanged; an empty list clears it
* - object_permission: Optional[LiteLLM_ObjectPermissionBase] - Customer-specific object permissions to control access to resources.
* Supported fields:
* * mcp_servers: List[str] - List of allowed MCP server IDs
* * mcp_access_groups: List[str] - List of MCP access group names
* * mcp_tool_permissions: Dict[str, List[str]] - Map of server ID to allowed tool names
* * vector_stores: List[str] - List of allowed vector store IDs
* * agents: List[str] - List of allowed agent IDs
* * agent_access_groups: List[str] - List of agent access group names
* Example: {"mcp_servers": ["server_1"], "vector_stores": ["vector_store_1"]}
* IF null or {} then no object-level restrictions apply.
*
* Example curl:
* ```
* curl --location 'http://0.0.0.0:4000/customer/update' --header "Authorization: Bearer $LITELLM_MASTER_KEY" --header 'Content-Type: application/json' --data '{
* "user_id": "test-litellm-user-4",
* "budget_id": "paid_tier",
* "models": ["gpt-4o-mini"]
* }'
*
* # Updating object permissions
* curl -L -X POST 'http://localhost:4000/customer/update' --header "Authorization: Bearer $LITELLM_MASTER_KEY" --header 'Content-Type: application/json' --data '{
* "user_id": "user_1",
* "object_permission": {
* "mcp_servers": ["server_3"],
* "vector_stores": ["vector_store_2", "vector_store_3"]
* }
* }'
*
* See below for all params
* ```
*/
post: operations["update_end_user_end_user_update_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/engines/{model}/chat/completions": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Chat Completion
* @description Follows the exact same API spec as `OpenAI's Chat API https://platform.openai.com/docs/api-reference/chat`
*
* ```bash
* curl -X POST http://localhost:4000/v1/chat/completions
* -H "Content-Type: application/json"
* -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* -d '{
* "model": "gpt-4o",
* "messages": [
* {
* "role": "user",
* "content": "Hello!"
* }
* ]
* }'
* ```
*/
post: operations["chat_completion_engines__model__chat_completions_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/engines/{model}/completions": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Completion
* @description Follows the exact same API spec as `OpenAI's Completions API https://platform.openai.com/docs/api-reference/completions`
*
* ```bash
* curl -X POST http://localhost:4000/v1/completions
* -H "Content-Type: application/json"
* -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* -d '{
* "model": "gpt-3.5-turbo-instruct",
* "prompt": "Once upon a time",
* "max_tokens": 50,
* "temperature": 0.7
* }'
* ```
*/
post: operations["completion_engines__model__completions_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/engines/{model}/embeddings": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Embeddings
* @description Follows the exact same API spec as `OpenAI's Embeddings API https://platform.openai.com/docs/api-reference/embeddings`
*
* ```bash
* curl -X POST http://localhost:4000/v1/embeddings
* -H "Content-Type: application/json"
* -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* -d '{
* "model": "text-embedding-ada-002",
* "input": "The quick brown fox jumps over the lazy dog"
* }'
* ```
*/
post: operations["embeddings_engines__model__embeddings_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/eu.assemblyai/{endpoint}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Assemblyai Proxy Route */
get: operations["assemblyai_proxy_route_eu_assemblyai__endpoint__get"];
/** Assemblyai Proxy Route */
put: operations["assemblyai_proxy_route_eu_assemblyai__endpoint__put"];
/** Assemblyai Proxy Route */
post: operations["assemblyai_proxy_route_eu_assemblyai__endpoint__post"];
/** Assemblyai Proxy Route */
delete: operations["assemblyai_proxy_route_eu_assemblyai__endpoint__delete"];
options?: never;
head?: never;
/** Assemblyai Proxy Route */
patch: operations["assemblyai_proxy_route_eu_assemblyai__endpoint__patch"];
trace?: never;
};
"/extraction": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Extract typed fields from a state
* @description Read a `state` text and up to 4 images and get one typed answer per named field in `fields`.
*
* Field types are `text`, `number`, `date`, `time`, `datetime`, `duration`, `list` (set `items`), `estimate`
* (set `min` and `max`) and `forecast` (set `min`, `max` and `horizon`). Each answer carries its `value`, which is
* null when the input does not determine it, plus `confidence`, `p_null`, text `span`s into the state and, for
* estimates, `mean`, `median`, `p10` and `p90`. Images are data URLs or `{content_type, base64}` objects, remote
* URLs are not accepted. An unknown field type, a field name that is not 1 to 64 letters, digits, `_`, `.` or `-`
* starting with a letter or `_`, a remote image URL or more than 4 images is rejected with a 400 before any
* provider is called. Any other top-level key, `stream: true` and OpenAI generation settings such as `temperature`
* included, is rejected with a 400 naming it unless `drop_params` or the deployment's `additional_drop_params`
* drops it. Settings inside `fields` and `options` go to the provider as sent, so a field missing a setting its
* type needs or an empty `state` with no images comes back with the provider's own error. `model` is any
* extraction model in the proxy model list, such as Cloudflare Clairic. The request and the answers keep the
* provider's format.
*/
post: operations["extraction_extraction_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/fal_ai/{endpoint}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Fal Ai Proxy Route */
get: operations["fal_ai_proxy_route_fal_ai__endpoint__get"];
/** Fal Ai Proxy Route */
put: operations["fal_ai_proxy_route_fal_ai__endpoint__put"];
/** Fal Ai Proxy Route */
post: operations["fal_ai_proxy_route_fal_ai__endpoint__post"];
/** Fal Ai Proxy Route */
delete: operations["fal_ai_proxy_route_fal_ai__endpoint__delete"];
options?: never;
head?: never;
/** Fal Ai Proxy Route */
patch: operations["fal_ai_proxy_route_fal_ai__endpoint__patch"];
trace?: never;
};
"/fallback": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Create Fallback
* @description Create or update fallbacks for a specific model.
*
* This endpoint allows you to configure fallback models separately from the general config.
* Fallbacks are triggered when a model call fails after retries.
*
* **Example Request:**
* ```json
* {
* "model": "gpt-3.5-turbo",
* "fallback_models": ["gpt-4", "claude-3-haiku"],
* "fallback_type": "general"
* }
* ```
*
* **Fallback Types:**
* - `general`: Standard fallbacks for any error (default)
* - `context_window`: Fallbacks specifically for context window exceeded errors
* - `content_policy`: Fallbacks specifically for content policy violations
*/
post: operations["create_fallback_fallback_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/fallback/login": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Fallback Login
* @description Create Proxy API Keys using Google Workspace SSO. Requires setting PROXY_BASE_URL in .env
* PROXY_BASE_URL should be the your deployed proxy endpoint, e.g. PROXY_BASE_URL="https://litellm-production-7002.up.railway.app/"
* Example:
*/
get: operations["fallback_login_fallback_login_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/fallback/{model}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Fallback
* @description Get fallback configuration for a specific model.
*
* **Parameters:**
* - `model`: The model name to get fallbacks for
* - `fallback_type`: Type of fallback to retrieve (query parameter)
*
* **Example:**
* ```
* GET /fallback/gpt-3.5-turbo?fallback_type=general
* ```
*/
get: operations["get_fallback_fallback__model__get"];
put?: never;
post?: never;
/**
* Delete Fallback
* @description Delete fallback configuration for a specific model.
*
* **Parameters:**
* - `model`: The model name to delete fallbacks for
* - `fallback_type`: Type of fallback to delete (query parameter)
*
* **Example:**
* ```
* DELETE /fallback/gpt-3.5-turbo?fallback_type=general
* ```
*/
delete: operations["delete_fallback_fallback__model__delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/files": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List Files
* @description Returns information about a specific file. that can be used across - Assistants API, Batch API
* This is the equivalent of GET https://api.openai.com/v1/files/
*
* Supports Identical Params as: https://platform.openai.com/docs/api-reference/files/list
*
* Example Curl
* ```
* curl http://localhost:4000/v1/files -H "Authorization: Bearer $LITELLM_MASTER_KEY"
*
* ```
*/
get: operations["list_files_files_get"];
put?: never;
/**
* Create File
* @description Upload a file that can be used across - Assistants API, Batch API
* This is the equivalent of POST https://api.openai.com/v1/files
*
* Supports Identical Params as: https://platform.openai.com/docs/api-reference/files/create
*
* Example Curl
* ```
* curl http://localhost:4000/v1/files -H "Authorization: Bearer $LITELLM_MASTER_KEY" -F purpose="batch" -F file="@mydata.jsonl"
* -F expires_after[anchor]="created_at" -F expires_after[seconds]=2592000
* ```
*/
post: operations["create_file_files_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/files/{file_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get File
* @description Returns information about a specific file. that can be used across - Assistants API, Batch API
* This is the equivalent of GET https://api.openai.com/v1/files/{file_id}
*
* Supports Identical Params as: https://platform.openai.com/docs/api-reference/files/retrieve
*
* Example Curl
* ```
* curl http://localhost:4000/v1/files/file-abc123 -H "Authorization: Bearer $LITELLM_MASTER_KEY"
*
* ```
*/
get: operations["get_file_files__file_id__get"];
put?: never;
post?: never;
/**
* Delete File
* @description Deletes a specified file. that can be used across - Assistants API, Batch API
* This is the equivalent of DELETE https://api.openai.com/v1/files/{file_id}
*
* Supports Identical Params as: https://platform.openai.com/docs/api-reference/files/delete
*
* Example Curl
* ```
* curl http://localhost:4000/v1/files/file-abc123 -X DELETE -H "Authorization: Bearer $OPENAI_API_KEY"
*
* ```
*/
delete: operations["delete_file_files__file_id__delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/files/{file_id}/content": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get File Content
* @description Returns information about a specific file. that can be used across - Assistants API, Batch API
* This is the equivalent of GET https://api.openai.com/v1/files/{file_id}/content
*
* Supports Identical Params as: https://platform.openai.com/docs/api-reference/files/retrieve-contents
*
* Example Curl
* ```
* curl http://localhost:4000/v1/files/file-abc123/content -H "Authorization: Bearer $LITELLM_MASTER_KEY"
*
* ```
*/
get: operations["get_file_content_files__file_id__content_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/fine_tuning/jobs": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* ✨ (Enterprise) List Fine-Tuning Jobs
* @description Lists fine-tuning jobs for the organization.
* This is the equivalent of GET https://api.openai.com/v1/fine_tuning/jobs
*
* Supported Query Params:
* - `custom_llm_provider`: Name of the LiteLLM provider
* - `after`: Identifier for the last job from the previous pagination request.
* - `limit`: Number of fine-tuning jobs to retrieve (default is 20).
*/
get: operations["list_fine_tuning_jobs_fine_tuning_jobs_get"];
put?: never;
/**
* ✨ (Enterprise) Create Fine-Tuning Job
* @description Creates a fine-tuning job which begins the process of creating a new model from a given dataset.
* This is the equivalent of POST https://api.openai.com/v1/fine_tuning/jobs
*
* Supports Identical Params as: https://platform.openai.com/docs/api-reference/fine-tuning/create
*
* Example Curl:
* ```
* curl http://localhost:4000/v1/fine_tuning/jobs -H "Content-Type: application/json" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -d '{
* "model": "gpt-3.5-turbo",
* "training_file": "file-abc123",
* "hyperparameters": {
* "n_epochs": 4
* }
* }'
* ```
*/
post: operations["create_fine_tuning_job_fine_tuning_jobs_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/fine_tuning/jobs/{fine_tuning_job_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* ✨ (Enterprise) Retrieve Fine-Tuning Job
* @description Retrieves a fine-tuning job.
* This is the equivalent of GET https://api.openai.com/v1/fine_tuning/jobs/{fine_tuning_job_id}
*
* Supported Query Params:
* - `custom_llm_provider`: Name of the LiteLLM provider
* - `fine_tuning_job_id`: The ID of the fine-tuning job to retrieve.
*/
get: operations["retrieve_fine_tuning_job_fine_tuning_jobs__fine_tuning_job_id__get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/fine_tuning/jobs/{fine_tuning_job_id}/cancel": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* ✨ (Enterprise) Cancel Fine-Tuning Jobs
* @description Cancel a fine-tuning job.
*
* This is the equivalent of POST https://api.openai.com/v1/fine_tuning/jobs/{fine_tuning_job_id}/cancel
*
* Supported Query Params:
* - `custom_llm_provider`: Name of the LiteLLM provider
* - `fine_tuning_job_id`: The ID of the fine-tuning job to cancel.
*/
post: operations["cancel_fine_tuning_job_fine_tuning_jobs__fine_tuning_job_id__cancel_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/gateway/daily/activity": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Gateway Daily Activity
* @description Successful and failed gateway requests, counted at the ASGI edge.
*
* Deployment-wide: the underlying table has no per-key or per-user dimension,
* so this is admin-only.
*/
get: operations["get_gateway_daily_activity_gateway_daily_activity_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/gateway/errors/activity": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Request Error Activity
* @description Failed requests over time by HTTP status, and the keys, teams, users and
* model groups they land on. Deployment-wide, so admin-only.
*/
get: operations["get_request_error_activity_gateway_errors_activity_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/gemini/{endpoint}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Gemini Proxy Route
* @description [Docs](https://docs.litellm.ai/docs/pass_through/google_ai_studio)
*/
get: operations["gemini_proxy_route_gemini__endpoint__get"];
/**
* Gemini Proxy Route
* @description [Docs](https://docs.litellm.ai/docs/pass_through/google_ai_studio)
*/
put: operations["gemini_proxy_route_gemini__endpoint__put"];
/**
* Gemini Proxy Route
* @description [Docs](https://docs.litellm.ai/docs/pass_through/google_ai_studio)
*/
post: operations["gemini_proxy_route_gemini__endpoint__post"];
/**
* Gemini Proxy Route
* @description [Docs](https://docs.litellm.ai/docs/pass_through/google_ai_studio)
*/
delete: operations["gemini_proxy_route_gemini__endpoint__delete"];
options?: never;
head?: never;
/**
* Gemini Proxy Route
* @description [Docs](https://docs.litellm.ai/docs/pass_through/google_ai_studio)
*/
patch: operations["gemini_proxy_route_gemini__endpoint__patch"];
trace?: never;
};
"/get/allowed_ips": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get Allowed Ips */
get: operations["get_allowed_ips_get_allowed_ips_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/get/config/callbacks": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Config
* @description For Admin UI - allows admin to view config via UI
* # return the callbacks and the env variables for the callback
*/
get: operations["get_config_get_config_callbacks_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/get/default_team_settings": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Default Team Settings
* @description Get the default team parameters (litellm_settings.default_team_params).
* Returns a structured object with values and descriptions for UI display.
*/
get: operations["get_default_team_settings_get_default_team_settings_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/get/internal_user_settings": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Internal User Settings
* @description Get all SSO settings from the litellm_settings configuration.
* Returns a structured object with values and descriptions for UI display.
*/
get: operations["get_internal_user_settings_get_internal_user_settings_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/get/latest_release_info": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Latest Release Info
* @description Latest stable LiteLLM GitHub release with the PR count of its whole minor line (e.g. v1.105.0 + v1.105.1)
* split into new features, bug fixes and other updates.
* Returns null when GitHub can't be reached so the dashboard upgrade banner simply doesn't render.
*/
get: operations["latest_release_info_get_latest_release_info_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/get/mcp_semantic_filter_settings": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Mcp Semantic Filter Settings
* @description Get MCP semantic filter configuration.
* Returns current settings for semantic tool filtering.
*/
get: operations["get_mcp_semantic_filter_settings_get_mcp_semantic_filter_settings_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/get/mcp_tool_search_settings": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Mcp Tool Search Settings
* @description Get the `litellm_settings.mcp_tool_search` configuration used by the native `mcp_tool_search` virtual tool.
*/
get: operations["get_mcp_tool_search_settings_get_mcp_tool_search_settings_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/get/sso_settings": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Sso Settings
* @description Get all SSO configuration settings from the dedicated SSO table.
* Returns a structured object with values and descriptions for UI display.
*/
get: operations["get_sso_settings_get_sso_settings_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/get/ui_settings": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Ui Settings
* @description Get UI-specific configuration flags.
* All authenticated users can fetch these settings for client-side behavior.
*/
get: operations["get_ui_settings_get_ui_settings_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/get/ui_theme_settings": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Ui Theme Settings
* @description Get UI theme configuration from the litellm_settings.
* Returns current logo settings for UI customization.
*
* Note: This endpoint is public (no authentication required) so all users can see custom branding.
* Only the /update/ui_theme_settings endpoint requires authentication for admins to change settings.
*/
get: operations["get_ui_theme_settings_get_ui_theme_settings_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/get/user_banner": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get User Banner
* @description Get the admin-published dashboard banner.
* Readable by any authenticated user; rendered on every dashboard page.
*/
get: operations["get_user_banner_get_user_banner_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/get/websearch_interception_settings": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Websearch Interception Settings
* @description Get web search interception configuration.
*
* Returns the current settings plus their schema, for the Admin UI to render.
*/
get: operations["get_websearch_interception_settings_get_websearch_interception_settings_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/get_favicon": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Favicon
* @description Get custom favicon for the admin UI.
*/
get: operations["get_favicon_get_favicon_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/get_image": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Image
* @description Get logo to show on admin UI
*/
get: operations["get_image_get_image_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/get_logo_url": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Logo Url
* @description Get the current logo URL from environment.
*
* Only HTTP(S) URLs are returned — those are intended to be loaded
* directly by the browser from a public/internal CDN. Local file
* paths set via ``UI_LOGO_PATH`` are NOT returned: they are admin-
* only filesystem details, the dashboard falls back to ``/get_image``
* which serves the file only when it is a supported image. Without
* this filter, the unauthenticated endpoint would disclose internal
* hostnames or filesystem paths to any caller.
*/
get: operations["get_logo_url_get_logo_url_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/gigachat/{endpoint}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Gigachat Proxy Route
* @description [Docs](https://docs.litellm.ai/docs/pass_through/gigachat)
*/
get: operations["gigachat_proxy_route_gigachat__endpoint__get"];
/**
* Gigachat Proxy Route
* @description [Docs](https://docs.litellm.ai/docs/pass_through/gigachat)
*/
put: operations["gigachat_proxy_route_gigachat__endpoint__put"];
/**
* Gigachat Proxy Route
* @description [Docs](https://docs.litellm.ai/docs/pass_through/gigachat)
*/
post: operations["gigachat_proxy_route_gigachat__endpoint__post"];
/**
* Gigachat Proxy Route
* @description [Docs](https://docs.litellm.ai/docs/pass_through/gigachat)
*/
delete: operations["gigachat_proxy_route_gigachat__endpoint__delete"];
options?: never;
head?: never;
/**
* Gigachat Proxy Route
* @description [Docs](https://docs.litellm.ai/docs/pass_through/gigachat)
*/
patch: operations["gigachat_proxy_route_gigachat__endpoint__patch"];
trace?: never;
};
"/global/activity": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Global Activity
* @description Get number of API Requests, total tokens through proxy
*
* {
* "daily_data": [
* const chartdata = [
* {
* date: 'Jan 22',
* api_requests: 10,
* total_tokens: 2000
* },
* {
* date: 'Jan 23',
* api_requests: 10,
* total_tokens: 12
* },
* ],
* "sum_api_requests": 20,
* "sum_total_tokens": 2012
* }
*/
get: operations["get_global_activity_global_activity_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/global/activity/cache_hits": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Global Activity
* @description Cache activity for the Admin UI cache dashboard, aggregated per call_type:
* cache hits vs successful LLM API requests vs failed requests, plus totals
* for the stat cards and the available key-alias/model filter options.
*
* Admin callers see proxy-wide activity; all other callers see only their own
* spend logs and those of teams they may view, exactly like ``/spend/logs/ui``.
*/
get: operations["get_global_activity_global_activity_cache_hits_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/global/activity/exceptions": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Global Activity Exceptions
* @description Get number of API Requests, total tokens through proxy
*
* {
* "daily_data": [
* const chartdata = [
* {
* date: 'Jan 22',
* num_rate_limit_exceptions: 10,
* },
* {
* date: 'Jan 23',
* num_rate_limit_exceptions: 10,
* },
* ],
* "sum_api_exceptions": 20,
* }
*/
get: operations["get_global_activity_exceptions_global_activity_exceptions_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/global/activity/exceptions/deployment": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Global Activity Exceptions Per Deployment
* @description Get number of 429 errors - Grouped by deployment
*
* [
* {
* "deployment": "https://azure-us-east-1.openai.azure.com/",
* "daily_data": [
* const chartdata = [
* {
* date: 'Jan 22',
* num_rate_limit_exceptions: 10
* },
* {
* date: 'Jan 23',
* num_rate_limit_exceptions: 12
* },
* ],
* "sum_num_rate_limit_exceptions": 20,
*
* },
* {
* "deployment": "https://azure-us-east-1.openai.azure.com/",
* "daily_data": [
* const chartdata = [
* {
* date: 'Jan 22',
* num_rate_limit_exceptions: 10,
* },
* {
* date: 'Jan 23',
* num_rate_limit_exceptions: 12
* },
* ],
* "sum_num_rate_limit_exceptions": 20,
*
* },
* ]
*/
get: operations["get_global_activity_exceptions_per_deployment_global_activity_exceptions_deployment_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/global/activity/model": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Global Activity Model
* @description Get number of API Requests, total tokens through proxy - Grouped by MODEL
*
* [
* {
* "model": "gpt-4",
* "daily_data": [
* const chartdata = [
* {
* date: 'Jan 22',
* api_requests: 10,
* total_tokens: 2000
* },
* {
* date: 'Jan 23',
* api_requests: 10,
* total_tokens: 12
* },
* ],
* "sum_api_requests": 20,
* "sum_total_tokens": 2012
*
* },
* {
* "model": "azure/gpt-4-turbo",
* "daily_data": [
* const chartdata = [
* {
* date: 'Jan 22',
* api_requests: 10,
* total_tokens: 2000
* },
* {
* date: 'Jan 23',
* api_requests: 10,
* total_tokens: 12
* },
* ],
* "sum_api_requests": 20,
* "sum_total_tokens": 2012
*
* },
* ]
*/
get: operations["get_global_activity_model_global_activity_model_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/global/all_end_users": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Global View All End Users
* @description [BETA] This is a beta endpoint. It will change.
*
* Use this to just get all the unique `end_users`
*/
get: operations["global_view_all_end_users_global_all_end_users_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/global/spend": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Global Spend
* @description [BETA] This is a beta endpoint. It will change.
*
* View total spend across all proxy keys
*/
get: operations["global_spend_global_spend_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/global/spend/all_tag_names": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Global Get All Tag Names */
get: operations["global_get_all_tag_names_global_spend_all_tag_names_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/global/spend/end_users": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Global Spend End Users
* @description [BETA] This is a beta endpoint. It will change.
*
* Use this to get the top 'n' keys with the highest spend, ordered by spend.
*/
post: operations["global_spend_end_users_global_spend_end_users_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/global/spend/keys": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Global Spend Keys
* @description [BETA] This is a beta endpoint. It will change.
*
* Use this to get the top 'n' keys with the highest spend, ordered by spend.
*/
get: operations["global_spend_keys_global_spend_keys_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/global/spend/logs": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Global Spend Logs
* @description [BETA] This is a beta endpoint. It will change.
*
* Use this to get global spend (spend per day for last 30d). Admin-only endpoint
*
* More efficient implementation of /spend/logs, by creating a view over the spend logs table.
*/
get: operations["global_spend_logs_global_spend_logs_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/global/spend/models": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Global Spend Models
* @description [BETA] This is a beta endpoint. It will change.
*
* Use this to get the top 'n' models with the highest spend, ordered by spend.
*/
get: operations["global_spend_models_global_spend_models_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/global/spend/provider": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Global Spend Provider
* @description Get breakdown of spend per provider
* [
* {
* "provider": "Azure OpenAI",
* "spend": 20
* },
* {
* "provider": "OpenAI",
* "spend": 10
* },
* {
* "provider": "VertexAI",
* "spend": 30
* }
* ]
*/
get: operations["get_global_spend_provider_global_spend_provider_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/global/spend/refresh": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Global Spend Refresh
* @description ADMIN ONLY / MASTER KEY Only Endpoint
*
* Globally refresh spend MonthlyGlobalSpend view
*/
post: operations["global_spend_refresh_global_spend_refresh_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/global/spend/report": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Global Spend Report
* @description Get Daily Spend per Team, based on specific startTime and endTime. Per team, view usage by each key, model
* [
* {
* "group-by-day": "2024-05-10",
* "teams": [
* {
* "team_name": "team-1"
* "spend": 10,
* "keys": [
* "key": "1213",
* "usage": {
* "model-1": {
* "cost": 12.50,
* "input_tokens": 1000,
* "output_tokens": 5000,
* "requests": 100
* },
* "audio-modelname1": {
* "cost": 25.50,
* "seconds": 25,
* "requests": 50
* },
* }
* }
* ]
* ]
* }
*/
get: operations["get_global_spend_report_global_spend_report_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/global/spend/reset": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Global Spend Reset
* @description ADMIN ONLY / MASTER KEY Only Endpoint
*
* Globally reset spend for All API Keys and Teams, maintain LiteLLM_SpendLogs
*
* 1. LiteLLM_SpendLogs will maintain the logs on spend, no data gets deleted from there
* 2. LiteLLM_VerificationTokens spend will be set = 0
* 3. LiteLLM_TeamTable spend will be set = 0
*/
post: operations["global_spend_reset_global_spend_reset_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/global/spend/tags": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Global View Spend Tags
* @description LiteLLM Enterprise - View Spend Per Request Tag. Used by LiteLLM UI
*
* Example Request:
* ```
* curl -X GET "http://0.0.0.0:4000/spend/tags" -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*
* Spend with Start Date and End Date
* ```
* curl -X GET "http://0.0.0.0:4000/spend/tags?start_date=2022-01-01&end_date=2022-02-01" -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*/
get: operations["global_view_spend_tags_global_spend_tags_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/global/spend/teams": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Global Spend Per Team
* @description [BETA] This is a beta endpoint. It will change.
*
* Use this to get daily spend, grouped by `team_id` and `date`
*/
get: operations["global_spend_per_team_global_spend_teams_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/guardrails": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Create Guardrail
* @description Create a new guardrail
*
* 👉 [Guardrail docs](https://docs.litellm.ai/docs/proxy/guardrails/quick_start)
*
* Example Request:
* ```bash
* curl -X POST "http://localhost:4000/guardrails" \
* -H "Authorization: Bearer <your_api_key>" \
* -H "Content-Type: application/json" \
* -d '{
* "guardrail": {
* "guardrail_name": "my-bedrock-guard",
* "litellm_params": {
* "guardrail": "bedrock",
* "mode": "pre_call",
* "guardrailIdentifier": "ff6ujrregl1q",
* "guardrailVersion": "DRAFT",
* "default_on": true
* },
* "guardrail_info": {
* "description": "Bedrock content moderation guardrail"
* }
* }
* }'
* ```
*
* Example Response:
* ```json
* {
* "guardrail_id": "123e4567-e89b-12d3-a456-426614174000",
* "guardrail_name": "my-bedrock-guard",
* "litellm_params": {
* "guardrail": "bedrock",
* "mode": "pre_call",
* "guardrailIdentifier": "ff6ujrregl1q",
* "guardrailVersion": "DRAFT",
* "default_on": true
* },
* "guardrail_info": {
* "description": "Bedrock content moderation guardrail"
* },
* "created_at": "2023-11-09T12:34:56.789Z",
* "updated_at": "2023-11-09T12:34:56.789Z"
* }
* ```
*/
post: operations["create_guardrail_guardrails_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/guardrails/apply_guardrail": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Apply Guardrail
* @description Apply a guardrail to text input and return the processed result.
*
* This endpoint allows testing guardrails by applying them to custom text inputs.
*/
post: operations["apply_guardrail_guardrails_apply_guardrail_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/guardrails/list": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List Guardrails
* @description List the guardrails that are available on the proxy server
*
* 👉 [Guardrail docs](https://docs.litellm.ai/docs/proxy/guardrails/quick_start)
*
* Example Request:
* ```bash
* curl -X GET "http://localhost:4000/guardrails/list" -H "Authorization: Bearer <your_api_key>"
* ```
*
* Example Response:
* ```json
* {
* "guardrails": [
* {
* "guardrail_name": "bedrock-pre-guard",
* "guardrail_info": {
* "params": [
* {
* "name": "toxicity_score",
* "type": "float",
* "description": "Score between 0-1 indicating content toxicity level"
* },
* {
* "name": "pii_detection",
* "type": "boolean"
* }
* ]
* }
* }
* ]
* }
* ```
*/
get: operations["list_guardrails_guardrails_list_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/guardrails/register": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Register Guardrail
* @description Register a guardrail for onboarding (team submission).
*
* Accepts a guardrail config in the
* [Generic Guardrail API](https://docs.litellm.ai/docs/adding_provider/generic_guardrail_api) format.
* The submission is stored with status `pending_review` until an admin approves it.
*/
post: operations["register_guardrail_guardrails_register_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/guardrails/submissions": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List Guardrail Submissions
* @description List team guardrail submissions. Returns only guardrails with a team_id.
*
* Admins see all submissions. Non-admin users see submissions for teams they are
* a member of.
*
* Status values: pending_review (team-registered, awaiting approval), active (approved), rejected.
*
* Optional filters:
* - status: pending_review | active | rejected
* - team_id: filter by specific team (non-admins must be a member of that team)
* - search: name/description
*/
get: operations["list_guardrail_submissions_guardrails_submissions_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/guardrails/submissions/{guardrail_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Guardrail Submission
* @description Get a single guardrail submission by id. Non-admins may only access submissions for teams they belong to.
*/
get: operations["get_guardrail_submission_guardrails_submissions__guardrail_id__get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/guardrails/submissions/{guardrail_id}/approve": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Approve Guardrail Submission
* @description Approve a pending guardrail submission: set status to active and initialize in memory (admin only).
*/
post: operations["approve_guardrail_submission_guardrails_submissions__guardrail_id__approve_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/guardrails/submissions/{guardrail_id}/reject": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Reject Guardrail Submission
* @description Reject a guardrail submission (admin only).
*/
post: operations["reject_guardrail_submission_guardrails_submissions__guardrail_id__reject_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/guardrails/test_custom_code": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Test Custom Code Guardrail
* @description Test custom code guardrail logic without creating a guardrail.
*
* This endpoint allows admins to experiment with custom code guardrails by:
* 1. Compiling the provided code in a sandbox
* 2. Executing the apply_guardrail function with test input
* 3. Returning the result (allow/block/modify)
*
* 👉 [Custom Code Guardrail docs](https://docs.litellm.ai/docs/proxy/guardrails/custom_code_guardrail)
*
* Example Request:
* ```bash
* curl -X POST "http://localhost:4000/guardrails/test_custom_code" \
* -H "Authorization: Bearer <your_api_key>" \
* -H "Content-Type: application/json" \
* -d '{
* "custom_code": "def apply_guardrail(inputs, request_data, input_type):\n for text in inputs[\"texts\"]:\n if regex_match(text, r\"\\d{3}-\\d{2}-\\d{4}\"):\n return block(\"SSN detected\")\n return allow()",
* "test_input": {
* "texts": ["My SSN is 123-45-6789"]
* },
* "input_type": "request"
* }'
* ```
*
* Example Success Response (blocked):
* ```json
* {
* "success": true,
* "result": {
* "action": "block",
* "reason": "SSN detected"
* },
* "error": null,
* "error_type": null
* }
* ```
*
* Example Success Response (allowed):
* ```json
* {
* "success": true,
* "result": {
* "action": "allow"
* },
* "error": null,
* "error_type": null
* }
* ```
*
* Example Success Response (modified):
* ```json
* {
* "success": true,
* "result": {
* "action": "modify",
* "texts": ["My SSN is [REDACTED]"]
* },
* "error": null,
* "error_type": null
* }
* ```
*
* Example Error Response (compilation error):
* ```json
* {
* "success": false,
* "result": null,
* "error": "Syntax error in custom code: invalid syntax (<guardrail>, line 1)",
* "error_type": "compilation"
* }
* ```
*/
post: operations["test_custom_code_guardrail_guardrails_test_custom_code_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/guardrails/ui/add_guardrail_settings": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Guardrail Ui Settings
* @description Get the UI settings for the guardrails
*
* Returns:
* - Supported entities for guardrails
* - Supported modes for guardrails
* - PII entity categories for UI organization
* - Content filter settings (patterns and categories)
*/
get: operations["get_guardrail_ui_settings_guardrails_ui_add_guardrail_settings_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/guardrails/ui/category_yaml/{category_name}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Category Yaml
* @description Get the YAML or JSON content for a specific content filter category.
*
* Args:
* category_name: The name of the category (e.g., "bias_gender", "harmful_self_harm")
*
* Returns:
* The raw YAML or JSON content of the category file with file type indicator
*/
get: operations["get_category_yaml_guardrails_ui_category_yaml__category_name__get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/guardrails/ui/major_airlines": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Major Airlines
* @description Get the major airlines list from IATA (competitor intent, airline type).
* Returns airline id, match variants (pipe-separated), and tags.
*/
get: operations["get_major_airlines_guardrails_ui_major_airlines_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/guardrails/ui/provider_specific_params": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Provider Specific Params
* @description Get provider-specific parameters for different guardrail types.
*
* Returns a dictionary mapping guardrail providers to their specific parameters,
* including parameter names, descriptions, and whether they are required.
*
* Example Response:
* ```json
* {
* "bedrock": {
* "guardrailIdentifier": {
* "description": "The ID of your guardrail on Bedrock",
* "required": true,
* "type": null
* },
* "guardrailVersion": {
* "description": "The version of your Bedrock guardrail (e.g., DRAFT or version number)",
* "required": true,
* "type": null
* }
* },
* "azure_content_safety_text_moderation": {
* "api_key": {
* "description": "API key for the Azure Content Safety Text Moderation guardrail",
* "required": false,
* "type": null
* },
* "optional_params": {
* "description": "Optional parameters for the Azure Content Safety Text Moderation guardrail",
* "required": true,
* "type": "nested",
* "fields": {
* "severity_threshold": {
* "description": "Severity threshold for the Azure Content Safety Text Moderation guardrail across all categories",
* "required": false,
* "type": null
* },
* "categories": {
* "description": "Categories to scan for the Azure Content Safety Text Moderation guardrail",
* "required": false,
* "type": "multiselect",
* "options": ["Hate", "SelfHarm", "Sexual", "Violence"],
* "default_value": None
* }
* }
* }
* }
* }
* ```
*/
get: operations["get_provider_specific_params_guardrails_ui_provider_specific_params_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/guardrails/usage/detail/{guardrail_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Guardrails Usage Detail
* @description Return single guardrail usage metrics and time series.
*/
get: operations["guardrails_usage_detail_guardrails_usage_detail__guardrail_id__get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/guardrails/usage/logs": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Guardrails Usage Logs
* @description Return paginated run logs for a guardrail (or policy) from SpendLogs via index.
*/
get: operations["guardrails_usage_logs_guardrails_usage_logs_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/guardrails/usage/overview": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Guardrails Usage Overview
* @description Return guardrail performance overview for the dashboard.
*/
get: operations["guardrails_usage_overview_guardrails_usage_overview_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/guardrails/validate_blocked_words_file": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Validate Blocked Words File
* @description Validate a blocked_words YAML file content.
*
* Args:
* request: Dictionary with 'file_content' key containing the YAML string
*
* Returns:
* Dictionary with 'valid' boolean and either 'message'/'errors' depending on result
*
* Example Request:
* ```json
* {
* "file_content": "blocked_words:\n - keyword: \"test\"\n action: \"BLOCK\""
* }
* ```
*
* Example Success Response:
* ```json
* {
* "valid": true,
* "message": "Valid YAML file with 2 blocked words"
* }
* ```
*
* Example Error Response:
* ```json
* {
* "valid": false,
* "errors": ["Entry 0: missing 'action' field"]
* }
* ```
*/
post: operations["validate_blocked_words_file_guardrails_validate_blocked_words_file_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/guardrails/{guardrail_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Guardrail Info
* @description Get detailed information about a specific guardrail by ID
*
* 👉 [Guardrail docs](https://docs.litellm.ai/docs/proxy/guardrails/quick_start)
*
* Example Request:
* ```bash
* curl -X GET "http://localhost:4000/guardrails/123e4567-e89b-12d3-a456-426614174000/info" \
* -H "Authorization: Bearer <your_api_key>"
* ```
*
* Example Response:
* ```json
* {
* "guardrail_id": "123e4567-e89b-12d3-a456-426614174000",
* "guardrail_name": "my-bedrock-guard",
* "litellm_params": {
* "guardrail": "bedrock",
* "mode": "pre_call",
* "guardrailIdentifier": "ff6ujrregl1q",
* "guardrailVersion": "DRAFT",
* "default_on": true
* },
* "guardrail_info": {
* "description": "Bedrock content moderation guardrail"
* },
* "created_at": "2023-11-09T12:34:56.789Z",
* "updated_at": "2023-11-09T12:34:56.789Z"
* }
* ```
*/
get: operations["get_guardrail_info_guardrails__guardrail_id__get"];
/**
* Update Guardrail
* @description Update an existing guardrail
*
* 👉 [Guardrail docs](https://docs.litellm.ai/docs/proxy/guardrails/quick_start)
*
* Example Request:
* ```bash
* curl -X PUT "http://localhost:4000/guardrails/123e4567-e89b-12d3-a456-426614174000" \
* -H "Authorization: Bearer <your_api_key>" \
* -H "Content-Type: application/json" \
* -d '{
* "guardrail": {
* "guardrail_name": "updated-bedrock-guard",
* "litellm_params": {
* "guardrail": "bedrock",
* "mode": "pre_call",
* "guardrailIdentifier": "ff6ujrregl1q",
* "guardrailVersion": "1.0",
* "default_on": true
* },
* "guardrail_info": {
* "description": "Updated Bedrock content moderation guardrail"
* }
* }
* }'
* ```
*
* Example Response:
* ```json
* {
* "guardrail_id": "123e4567-e89b-12d3-a456-426614174000",
* "guardrail_name": "updated-bedrock-guard",
* "litellm_params": {
* "guardrail": "bedrock",
* "mode": "pre_call",
* "guardrailIdentifier": "ff6ujrregl1q",
* "guardrailVersion": "1.0",
* "default_on": true
* },
* "guardrail_info": {
* "description": "Updated Bedrock content moderation guardrail"
* },
* "created_at": "2023-11-09T12:34:56.789Z",
* "updated_at": "2023-11-09T13:45:12.345Z"
* }
* ```
*/
put: operations["update_guardrail_guardrails__guardrail_id__put"];
post?: never;
/**
* Delete Guardrail
* @description Delete a guardrail
*
* 👉 [Guardrail docs](https://docs.litellm.ai/docs/proxy/guardrails/quick_start)
*
* Example Request:
* ```bash
* curl -X DELETE "http://localhost:4000/guardrails/123e4567-e89b-12d3-a456-426614174000" \
* -H "Authorization: Bearer <your_api_key>"
* ```
*
* Example Response:
* ```json
* {
* "message": "Guardrail 123e4567-e89b-12d3-a456-426614174000 deleted successfully"
* }
* ```
*/
delete: operations["delete_guardrail_guardrails__guardrail_id__delete"];
options?: never;
head?: never;
/**
* Patch Guardrail
* @description Partially update an existing guardrail
*
* 👉 [Guardrail docs](https://docs.litellm.ai/docs/proxy/guardrails/quick_start)
*
* This endpoint allows updating specific fields of a guardrail without sending the entire object.
* Only the following fields can be updated:
* - guardrail_name: The name of the guardrail
* - default_on: Whether the guardrail is enabled by default
* - guardrail_info: Additional information about the guardrail
*
* Example Request:
* ```bash
* curl -X PATCH "http://localhost:4000/guardrails/123e4567-e89b-12d3-a456-426614174000" \
* -H "Authorization: Bearer <your_api_key>" \
* -H "Content-Type: application/json" \
* -d '{
* "guardrail_name": "updated-name",
* "default_on": true,
* "guardrail_info": {
* "description": "Updated description"
* }
* }'
* ```
*
* Example Response:
* ```json
* {
* "guardrail_id": "123e4567-e89b-12d3-a456-426614174000",
* "guardrail_name": "updated-name",
* "litellm_params": {
* "guardrail": "bedrock",
* "mode": "pre_call",
* "guardrailIdentifier": "ff6ujrregl1q",
* "guardrailVersion": "DRAFT",
* "default_on": true
* },
* "guardrail_info": {
* "description": "Updated description"
* },
* "created_at": "2023-11-09T12:34:56.789Z",
* "updated_at": "2023-11-09T14:22:33.456Z"
* }
* ```
*/
patch: operations["patch_guardrail_guardrails__guardrail_id__patch"];
trace?: never;
};
"/guardrails/{guardrail_id}/info": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Guardrail Info
* @description Get detailed information about a specific guardrail by ID
*
* 👉 [Guardrail docs](https://docs.litellm.ai/docs/proxy/guardrails/quick_start)
*
* Example Request:
* ```bash
* curl -X GET "http://localhost:4000/guardrails/123e4567-e89b-12d3-a456-426614174000/info" \
* -H "Authorization: Bearer <your_api_key>"
* ```
*
* Example Response:
* ```json
* {
* "guardrail_id": "123e4567-e89b-12d3-a456-426614174000",
* "guardrail_name": "my-bedrock-guard",
* "litellm_params": {
* "guardrail": "bedrock",
* "mode": "pre_call",
* "guardrailIdentifier": "ff6ujrregl1q",
* "guardrailVersion": "DRAFT",
* "default_on": true
* },
* "guardrail_info": {
* "description": "Bedrock content moderation guardrail"
* },
* "created_at": "2023-11-09T12:34:56.789Z",
* "updated_at": "2023-11-09T12:34:56.789Z"
* }
* ```
*/
get: operations["get_guardrail_info_guardrails__guardrail_id__info_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/health": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Health Endpoint
* @description 🚨 USE `/health/liveliness` to health check the proxy 🚨
*
* See more 👉 https://docs.litellm.ai/docs/proxy/health
*
*
* Check the health of all the endpoints in config.yaml
*
* To run health checks in the background, add this to config.yaml:
* ```
* general_settings:
* # ... other settings
* background_health_checks: True
* ```
* else, the health checks will be run on models when /health is called.
*
* To skip deployments that set ``model_info.disable_background_health_check: true``
* on ``GET /health`` as well as in the background loop, set
* ``general_settings.health_check_skip_disabled_background_models: true``.
*/
get: operations["health_endpoint_health_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/health/backlog": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Health Backlog
* @description Returns the number of HTTP requests currently in-flight on this uvicorn worker.
*
* Use this to measure per-pod queue depth. A high value means the worker is
* processing many concurrent requests — requests arriving now will have to wait
* for the event loop to get to them, adding latency before LiteLLM even starts
* its own timer.
*/
get: operations["health_backlog_health_backlog_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/health/deployment_warnings": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Deployment Warnings
* @description List the active deployment warnings with their severity and whether an admin ignored or snoozed them.
* Readable by proxy admins and proxy admin viewers.
*/
get: operations["get_deployment_warnings_health_deployment_warnings_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/health/deployment_warnings/{warning_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
/**
* Update Deployment Warning Ignore
* @description Ignore, snooze, or restore one deployment warning for every admin of this proxy.
* Only proxy admins can call this, and it needs a database.
*/
patch: operations["update_deployment_warning_ignore_health_deployment_warnings__warning_id__patch"];
trace?: never;
};
"/health/drain": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Health Drain
* @description Graceful-drain probe for Kubernetes ``preStop`` hooks.
*
* Disabled by default and returns 404 unless ``general_settings`` sets
* ``enable_drain_endpoint: true``. Calling it flips a process-wide
* shutting-down flag, so a successful call permanently takes the worker out
* of rotation until the pod restarts.
*
* Because the kubelet calls preStop hooks without proxy credentials, the
* endpoint does not require ``user_api_key_auth``. To prevent any
* pod-reachable caller from triggering shutdown, set
* ``general_settings.drain_endpoint_token`` (or the ``DRAIN_ENDPOINT_TOKEN``
* env var) and supply the same value on the ``X-Drain-Token`` header from
* the preStop hook. Calls without the header (or with a wrong value) get a
* 401 and have no side effect.
*
* When enabled, it marks the worker as shutting down (so /health/readiness
* and /health/liveliness immediately start returning 503, removing the pod
* from service) and blocks until the in-flight request counter drains to
* zero or ``GRACEFUL_SHUTDOWN_TIMEOUT`` elapses. Unlike a fixed ``sleep``,
* this returns as soon as real in-flight work is done.
*
* Wire it up as:
*
* ```yaml
* lifecycle:
* preStop:
* httpGet:
* path: /health/drain
* port: 4000
* httpHeaders:
* - name: X-Drain-Token
* value: <same value as drain_endpoint_token>
* ```
*/
get: operations["health_drain_health_drain_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/health/history": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Health Check History Endpoint
* @description Get health check history for models
*
* Returns historical health check data with optional filtering.
*/
get: operations["health_check_history_endpoint_health_history_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/health/latest": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Latest Health Checks Endpoint
* @description Get the latest health check status for all models
*
* Returns the most recent health check result for each model.
*/
get: operations["latest_health_checks_endpoint_health_latest_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/health/license": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Health License Endpoint
* @description Return metadata about the configured LiteLLM license without exposing the key.
*/
get: operations["health_license_endpoint_health_license_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/health/liveliness": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Health Liveliness
* @description Unprotected endpoint for checking if worker is alive.
*
* Returns 503 once graceful shutdown has begun so Kubernetes stops counting
* the draining pod as live and terminates it on schedule.
*/
get: operations["health_liveliness_health_liveliness_get"];
put?: never;
post?: never;
delete?: never;
/**
* Health Liveliness Options
* @description Options endpoint for health/liveliness check.
*/
options: operations["health_liveliness_options_health_liveliness_options"];
head?: never;
patch?: never;
trace?: never;
};
"/health/liveness": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Health Liveliness
* @description Unprotected endpoint for checking if worker is alive.
*
* Returns 503 once graceful shutdown has begun so Kubernetes stops counting
* the draining pod as live and terminates it on schedule.
*/
get: operations["health_liveliness_health_liveness_get"];
put?: never;
post?: never;
delete?: never;
/**
* Health Liveliness Options
* @description Options endpoint for health/liveliness check.
*/
options: operations["health_liveliness_options_health_liveness_options"];
head?: never;
patch?: never;
trace?: never;
};
"/health/readiness": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Health Readiness
* @description Public readiness probe. Returns a low-detail payload safe to expose to
* unauthenticated load balancers — `status` plus `db` so orchestrators and
* external probes can distinguish "healthy" from "DB unreachable" without a
* credential. Admins can opt into the legacy detailed payload with
* general_settings.allow_public_health_readiness_details.
*/
get: operations["health_readiness_health_readiness_get"];
put?: never;
post?: never;
delete?: never;
/**
* Health Readiness Options
* @description Options endpoint for health/readiness check.
*/
options: operations["health_readiness_options_health_readiness_options"];
head?: never;
patch?: never;
trace?: never;
};
"/health/readiness/details": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Health Readiness Details
* @description Authenticated readiness diagnostics with DB/cache/callback metadata.
*/
get: operations["health_readiness_details_health_readiness_details_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/health/services": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Health Services Endpoint
* @description Use this admin-only endpoint to check if the service is healthy.
*
* Example:
* ```
* curl -L -X GET 'http://0.0.0.0:4000/health/services?service=datadog' -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*/
get: operations["health_services_endpoint_health_services_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/health/shared-status": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Shared Health Check Status Endpoint
* @description Get the status of shared health check coordination across pods.
*
* Returns information about Redis connectivity, lock status, and cache status.
*/
get: operations["shared_health_check_status_endpoint_health_shared_status_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/health/test_connection": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Test Model Connection
* @description Test a direct connection to a specific model.
*
* This endpoint allows you to verify if your proxy can successfully connect to a specific model.
* It's useful for troubleshooting model connectivity issues without going through the full proxy routing.
*
* Example:
* ```bash
* # If model is configured in proxy_config.yaml, you only need to specify the model name:
* curl -X POST 'http://localhost:4000/health/test_connection' \
* -H "Authorization: Bearer $LITELLM_MASTER_KEY" \
* -H 'Content-Type: application/json' \
* -d '{
* "litellm_params": {
* "model": "gpt-4o"
* },
* "mode": "chat"
* }'
*
* # The endpoint will automatically use api_key, api_base, etc. from proxy_config.yaml
*
* # You can also override specific params or test with custom credentials:
* curl -X POST 'http://localhost:4000/health/test_connection' \
* -H "Authorization: Bearer $LITELLM_MASTER_KEY" \
* -H 'Content-Type: application/json' \
* -d '{
* "litellm_params": {
* "model": "azure/gpt-4o",
* "api_key": "os.environ/AZURE_OPENAI_API_KEY",
* "api_base": "os.environ/AZURE_OPENAI_ENDPOINT",
* "api_version": "2024-10-21"
* },
* "mode": "chat"
* }'
* ```
*
* Note:
* - If the model is configured in proxy_config.yaml, credentials (api_key, api_base, etc.)
* will be automatically loaded from the config (with resolved environment variables).
* - A request naming a stored credential (`litellm_credential_name`) that the configuration
* does not name is probed with that credential instead, and inherits no credentials
* from the configuration its model string happened to match.
* - You can override specific params by including them in the request.
* - You can use `os.environ/VARIABLE_NAME` syntax to reference environment variables,
* which will be resolved automatically (same as in proxy_config.yaml).
*
* Returns:
* dict: A dictionary containing the health check result with either success information or error details.
*/
post: operations["test_model_connection_health_test_connection_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/images/edits": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Image Edit Api
* @description Follows the OpenAI Images API spec: https://platform.openai.com/docs/api-reference/images/create
*
* ```bash
* curl -s -D >(grep -i x-request-id >&2) -o >(jq -r '.data[0].b64_json' | base64 --decode > gift-basket.png) -X POST "http://localhost:4000/v1/images/edits" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -F "model=gpt-image-1" -F "image[][email protected]" -F 'prompt=Create a studio ghibli image of this'
* ```
*/
post: operations["image_edit_api_images_edits_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/images/generations": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/** Image Generation */
post: operations["image_generation_images_generations_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/interactions": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Create Interaction
* @description Create a new interaction using Google's Interactions API.
*
* Per OpenAPI spec: POST /{api_version}/interactions
*
* Supports both model interactions and agent interactions:
* - Model: Provide `model` parameter (e.g., "gemini-2.5-flash")
* - Agent: Provide `agent` parameter (e.g., "deep-research-pro-preview-12-2025")
*
* Example:
* ```bash
* curl -X POST "http://localhost:4000/v1beta/interactions" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -H "Content-Type: application/json" -d '{
* "model": "gemini/gemini-2.5-flash",
* "input": "Hello, how are you?"
* }'
* ```
*/
post: operations["create_interaction_interactions_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/interactions/{interaction_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Interaction
* @description Get an interaction by ID.
*
* Per OpenAPI spec: GET /{api_version}/interactions/{interaction_id}
*/
get: operations["get_interaction_interactions__interaction_id__get"];
put?: never;
post?: never;
/**
* Delete Interaction
* @description Delete an interaction by ID.
*
* Per OpenAPI spec: DELETE /{api_version}/interactions/{interaction_id}
*/
delete: operations["delete_interaction_interactions__interaction_id__delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/interactions/{interaction_id}/cancel": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Cancel Interaction
* @description Cancel an interaction by ID.
*
* Per OpenAPI spec: POST /{api_version}/interactions/{interaction_id}/cancel
*/
post: operations["cancel_interaction_interactions__interaction_id__cancel_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/introspect": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Introspect Endpoint
* @description RFC 7662 introspection for gateway-issued session tokens (``llm_session_`` /
* ``llm_srefresh_``), so an external gateway can validate them without the signing
* secret. The caller authenticates with a LiteLLM virtual key (section 2.1, enforced by
* the route dependency); any token the gateway cannot vouch for answers
* ``{"active": false}`` with no further detail.
*/
post: operations["introspect_endpoint_introspect_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/invitation/delete": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Invitation Delete
* @description Delete invitation link
*
* ```
* curl -X POST 'http://localhost:4000/invitation/delete' -H 'Content-Type: application/json' -d '{
* "invitation_id": "1234" // 👈 id of invitation in 'LiteLLM_InvitationTable'
* }'
* ```
*/
post: operations["invitation_delete_invitation_delete_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/invitation/info": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Invitation Info
* @description Allow admin to create invite links, to onboard new users to Admin UI.
*
* ```
* curl -X POST 'http://localhost:4000/invitation/new' -H 'Content-Type: application/json' -d '{
* "user_id": "1234" // 👈 id of user in 'LiteLLM_UserTable'
* }'
* ```
*/
get: operations["invitation_info_invitation_info_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/invitation/new": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* New Invitation
* @description Allow admin to create invite links, to onboard new users to Admin UI.
*
* ```
* curl -X POST 'http://localhost:4000/invitation/new' -H 'Content-Type: application/json' -d '{
* "user_id": "1234" // 👈 id of user in 'LiteLLM_UserTable'
* }'
* ```
*/
post: operations["new_invitation_invitation_new_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/invitation/update": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Invitation Update
* @description Update when invitation is accepted
*
* ```
* curl -X POST 'http://localhost:4000/invitation/update' -H 'Content-Type: application/json' -d '{
* "invitation_id": "1234" // 👈 id of invitation in 'LiteLLM_InvitationTable'
* "is_accepted": True // when invitation is accepted
* }'
* ```
*/
post: operations["invitation_update_invitation_update_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/jwt/key/mapping/delete": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/** Delete Jwt Key Mapping */
post: operations["delete_jwt_key_mapping_jwt_key_mapping_delete_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/jwt/key/mapping/info": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Info Jwt Key Mapping */
get: operations["info_jwt_key_mapping_jwt_key_mapping_info_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/jwt/key/mapping/list": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** List Jwt Key Mappings */
get: operations["list_jwt_key_mappings_jwt_key_mapping_list_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/jwt/key/mapping/new": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/** Create Jwt Key Mapping */
post: operations["create_jwt_key_mapping_jwt_key_mapping_new_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/jwt/key/mapping/update": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/** Update Jwt Key Mapping */
post: operations["update_jwt_key_mapping_jwt_key_mapping_update_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/key/aliases": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Key Aliases
* @description Lists key aliases with pagination and optional search.
*
* Non-admin users only see aliases for keys they own or keys belonging to
* their teams.
*
* Returns:
* {
* "aliases": List[str],
* "total_count": int,
* "current_page": int,
* "total_pages": int,
* "size": int,
* }
*/
get: operations["key_aliases_key_aliases_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/key/block": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Block Key
* @description Block an Virtual key from making any requests.
*
* Parameters:
* - key: str - The key to block. Can be either the unhashed key (sk-...) or the hashed key value
*
* Example:
* ```bash
* curl --location 'http://0.0.0.0:4000/key/block' --header "Authorization: Bearer $LITELLM_MASTER_KEY" --header 'Content-Type: application/json' --data '{
* "key": "sk-Fn8Ej39NxjAXrvpUGKghGw"
* }'
* ```
*
* Note: This is an admin-only endpoint. Only proxy admins, team admins, or org admins can block keys.
*/
post: operations["block_key_key_block_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/key/bulk_update": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Bulk Update Keys
* @description Bulk update multiple keys at once.
*
* This endpoint allows updating multiple keys in a single request. Each key update
* is processed independently - if some updates fail, others will still succeed.
*
* Parameters:
* - keys: List[BulkUpdateKeyRequestItem] - List of key update requests, each containing:
* - key: str - The key identifier (token) to update
* - budget_id: Optional[str] - Budget ID associated with the key
* - max_budget: Optional[float] - Max budget for key
* - team_id: Optional[str] - Team ID associated with key
* - tags: Optional[List[str]] - Tags for organizing keys
* - object_permission: Optional[LiteLLM_ObjectPermissionBase] - key-specific object permission, as on /key/update
*
* Only the fields an item carries are written: a field left out keeps its current value, and a field
* sent explicitly, null included, is applied exactly as /key/update applies it.
*
* Returns:
* - total_requested: int - Total number of keys requested for update
* - successful_updates: List[SuccessfulKeyUpdate] - List of successfully updated keys with their updated info
* - failed_updates: List[FailedKeyUpdate] - List of failed updates with key_info and failed_reason
*
* Example request:
* ```bash
* curl --location 'http://0.0.0.0:4000/key/bulk_update' --header "Authorization: Bearer $LITELLM_MASTER_KEY" --header 'Content-Type: application/json' --data '{
* "keys": [
* {
* "key": "sk-<your-virtual-key>",
* "max_budget": 100.0,
* "team_id": "team-123",
* "tags": ["production", "api"]
* },
* {
* "key": "sk-5678",
* "budget_id": "budget-456",
* "tags": ["staging"]
* }
* ]
* }'
* ```
*/
post: operations["bulk_update_keys_key_bulk_update_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/key/delete": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Delete Key Fn
* @description Delete a key from the key management system.
*
* Parameters::
* - keys (List[str]): A list of keys or hashed keys to delete. Example {"keys": ["sk-QWrxEynunsNpV1zT48HIrw", "837e17519f44683334df5291321d97b8bf1098cd490e49e215f6fea935aa28be"]}
* - key_aliases (List[str]): A list of key aliases to delete. Can be passed instead of `keys`.Example {"key_aliases": ["alias1", "alias2"]}
*
* Returns:
* - deleted_keys (List[str]): A list of deleted keys. Example {"deleted_keys": ["sk-QWrxEynunsNpV1zT48HIrw", "837e17519f44683334df5291321d97b8bf1098cd490e49e215f6fea935aa28be"]}
*
* Example:
* ```bash
* curl --location 'http://0.0.0.0:4000/key/delete' --header "Authorization: Bearer $LITELLM_MASTER_KEY" --header 'Content-Type: application/json' --data '{
* "keys": ["sk-QWrxEynunsNpV1zT48HIrw"]
* }'
* ```
*
* Raises:
* HTTPException: If an error occurs during key deletion.
*/
post: operations["delete_key_fn_key_delete_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/key/generate": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Generate Key Fn
* @description Generate an API key based on the provided data.
*
* Docs: https://docs.litellm.ai/docs/proxy/virtual_keys
*
* Parameters:
* - duration: Optional[str] - Specify the length of time the token is valid for. You can set duration as seconds ("30s"), minutes ("30m"), hours ("30h"), days ("30d").
* - key_alias: Optional[str] - User defined key alias
* - key: Optional[str] - User defined key value. Must start with 'sk-' and be at least 16 characters long. If not set, a 16-digit unique sk-key is created for you.
* - team_id: Optional[str] - The team id of the key
* - user_id: Optional[str] - The user id of the key. Non-admin callers may only pass their own id, except a team admin creating a key in a team they administer for another member of that team, when a proxy admin has added `create_member_keys` to `team_admin_editable_team_fields`. That member must have the `internal_user` or `internal_user_viewer` role, and the key is always an `llm_api` key.
* - agent_id: Optional[str] - The agent id associated with the key.
* - organization_id: Optional[str] - The organization id of the key. If not set, and team_id is set, the organization id will be the same as the team id. If conflict, an error will be raised.
* - project_id: Optional[str] - The project id of the key. When set, models and max_budget are validated against the project's limits.
* - budget_id: Optional[str] - The budget id associated with the key. Created by calling `/budget/new`.
* - end_user_budget_id: Optional[str] - Proxy admin only. Budget id applied to end users first seen through this key that carry no budget of their own. Takes precedence over `litellm_settings.max_end_user_budget_id`.
* - models: Optional[list] - Model_name's a user is allowed to call. (if empty, key is allowed to call all models)
* - aliases: Optional[dict] - Any alias mappings, on top of anything in the config.yaml model list. - https://docs.litellm.ai/docs/proxy/virtual_keys#managing-auth---upgradedowngrade-models
* - config: Optional[dict] - any key-specific configs, overrides config in config.yaml
* - spend: Optional[int] - Amount spent by key. Default is 0. Will be updated by proxy whenever key is used. https://docs.litellm.ai/docs/proxy/virtual_keys#managing-auth---tracking-spend
* - send_invite_email: Optional[bool] - Whether to send an invite email to the user_id, with the generate key
* - max_budget: Optional[float] - Specify max budget for a given key.
* - budget_duration: Optional[str] - Budget is reset at the end of specified duration. If not set, budget is never reset. You can set duration as seconds ("30s"), minutes ("30m"), hours ("30h"), days ("30d").
* - max_parallel_requests: Optional[int] - Rate limit a user based on the number of parallel requests. Raises 429 error, if user's parallel requests > x.
* - metadata: Optional[dict] - Metadata for key, store information for key. Example metadata = {"team": "core-infra", "app": "app2", "email": "[email protected]" }
* - guardrails: Optional[List[str]] - List of active guardrails for the key
* - policies: Optional[List[str]] - List of policy names to apply to the key. Policies define guardrails, conditions, and inheritance rules.
* - disable_global_guardrails: Optional[bool] - Whether to disable global guardrails for the key. Proxy admin only.
* - throttle_on_budget_exceeded: Optional[bool] - When the key exceeds its max_budget, throttle its tpm/rpm to the global budget_exceeded_throttle_percentage instead of blocking the key entirely.
* - enable_prompt_caching: Optional[bool] - Auto-inject prompt caching breakpoints (Anthropic cache_control markers) on requests made with this key. Supported Claude models on Anthropic, Bedrock, Vertex AI, and Azure AI only.
* - permissions: Optional[dict] - key-specific permissions. Currently just used for turning off pii masking (if connected). Example - {"pii": false}
* - model_max_budget: Optional[Dict[str, BudgetConfig]] - Model-specific budgets {"gpt-4": {"budget_limit": 0.0005, "time_period": "30d"}}}. IF null or {} then no model specific budget.
* - budget_fallbacks: Optional[Dict[str, List[str]]] - Per-model fallback chain tried in order when that model's own `model_max_budget` is exceeded, e.g. {"gpt-4o": ["gpt-4o-mini"]}.
* - model_rpm_limit: Optional[dict] - key-specific model rpm limit. Example - {"text-davinci-002": 1000, "gpt-3.5-turbo": 1000}. IF null or {} then no model specific rpm limit.
* - model_tpm_limit: Optional[dict] - key-specific model tpm limit. Example - {"text-davinci-002": 1000, "gpt-3.5-turbo": 1000}. IF null or {} then no model specific tpm limit.
* - default_estimated_output_tokens: Optional[int] - Proxy admin only. Expected output tokens reserved for TPM limiting when a request omits max_tokens. Positive integer. Falls back to the team setting, then to the built-in estimate.
* - default_estimated_output_tokens_per_model: Optional[dict] - Proxy admin only. Per-model override of the above. Example - {"gpt-4": 4096, "gpt-3.5-turbo": 1024}. Takes precedence over the key-wide value.
* - mcp_rpm_limit: Optional[dict] - key-specific per-MCP-server rpm limit, keyed by MCP server name (alias if set, else the configured name). Example - {"github": 100, "slack": 200}. IF null or {} then no MCP-specific rpm limit.
* - tag_rpm_limit: Optional[dict] - key-specific per-request-tag rpm limit, keyed by request tag. Example - {"cell-1": 1000, "cell-2": 500}. Each tag gets an independent counter; requests whose tag is absent fall back to the key-level rpm limit.
* - tpm_limit_type: Optional[str] - Type of tpm limit. Options: "best_effort_throughput" (no error if we're overallocating tpm), "guaranteed_throughput" (raise an error if we're overallocating tpm), "dynamic" (dynamically exceed limit when no 429 errors). Defaults to "best_effort_throughput".
* - rpm_limit_type: Optional[str] - Type of rpm limit. Options: "best_effort_throughput" (no error if we're overallocating rpm), "guaranteed_throughput" (raise an error if we're overallocating rpm), "dynamic" (dynamically exceed limit when no 429 errors). Defaults to "best_effort_throughput".
* - allowed_cache_controls: Optional[list] - List of allowed cache control values. Example - ["no-cache", "no-store"]. See all values - https://docs.litellm.ai/docs/proxy/caching#turn-on--off-caching-per-request
* - blocked: Optional[bool] - Whether the key is blocked.
* - rpm_limit: Optional[int] - Specify rpm limit for a given key (Requests per minute)
* - tpm_limit: Optional[int] - Specify tpm limit for a given key (Tokens per minute)
* - tpd_limit: Optional[int] - Specify tpd limit for a given key (Tokens per day). Charged by batch submissions instead of tpm_limit/rpm_limit.
* - soft_budget: Optional[float] - Specify soft budget for a given key. Will trigger a slack alert when this soft budget is reached.
* - tags: Optional[List[str]] - Tags for [tracking spend](https://litellm.vercel.app/docs/proxy/enterprise#tracking-spend-for-custom-tags) and/or doing [tag-based routing](https://litellm.vercel.app/docs/proxy/tag_routing).
* - prompts: Optional[List[str]] - List of prompts that the key is allowed to use.
* - enforced_params: Optional[List[str]] - List of enforced params for the key (Enterprise only). [Docs](https://docs.litellm.ai/docs/proxy/enterprise#enforce-required-params-for-llm-requests)
* - prompts: Optional[List[str]] - List of prompts that the key is allowed to use.
* - allowed_routes: Optional[list] - List of allowed routes for the key. Store the actual route or store a wildcard pattern for a set of routes. Example - ["/chat/completions", "/embeddings", "/keys/*"]
* - allowed_passthrough_routes: Optional[list] - List of allowed pass through endpoints for the key. Store the actual endpoint or store a wildcard pattern for a set of endpoints. Example - ["/my-custom-endpoint"]. Use this instead of allowed_routes, if you just want to specify which pass through endpoints the key can access, without specifying the routes. If allowed_routes is specified, allowed_pass_through_endpoints is ignored.
* - denied_passthrough_routes: Optional[list] - List of pass through routes the key may not call, even if allowed by `allowed_passthrough_routes` or `allowed_routes`. Matches exact paths, path prefixes, and trailing `*` wildcards. Applies together with the team's `denied_passthrough_routes`. Example - ["/my-custom-endpoint/admin"].
* - object_permission: Optional[LiteLLM_ObjectPermissionBase] - key-specific object permission. Example - {"vector_stores": ["vector_store_1", "vector_store_2"], "agents": ["agent_1", "agent_2"], "agent_access_groups": ["dev_group"]}. IF null or {} then no object permission.
* - key_type: Optional[str] - Type of key that determines default allowed routes. Options: "llm_api" (can call LLM API routes), "management" (can call management routes), "read_only" (can only call info/read routes), "default" (Full Access in the Admin UI, skips `default_key_generate_params.allowed_routes`). When omitted, the key gets `default_key_generate_params.allowed_routes` if that is set and the request has no `allowed_routes`, else it works like "default".
* - prompts: Optional[List[str]] - List of allowed prompts for the key. If specified, the key will only be able to use these specific prompts.
* - auto_rotate: Optional[bool] - Whether this key should be automatically rotated (regenerated)
* - rotation_interval: Optional[str] - How often to auto-rotate this key (e.g., '30s', '30m', '30h', '30d'). Required if auto_rotate=True.
* - allowed_vector_store_indexes: Optional[List[dict]] - List of allowed vector store indexes for the key. Example - [{"index_name": "my-index", "index_permissions": ["write", "read"]}]. If specified, the key will only be able to use these specific vector store indexes. Create index, using `/v1/indexes` endpoint.
* - router_settings: Optional[UpdateRouterConfig] - key-specific router settings. Example - {"model_group_retry_policy": {"gpt-4": {"RateLimitErrorRetries": 5}}}. IF null or {} then no router settings.
* - access_group_ids: Optional[List[str]] - List of access group IDs to associate with the key. Access groups define which models a key can access. Example - ["access_group_1", "access_group_2"].
* - budget_limits: Optional[list] - List of concurrent budget windows for the key. Each window specifies a budget_limit, time_period, and optional budget_duration. Example - [{"budget_limit": 10.0, "time_period": "1d"}, {"budget_limit": 50.0, "time_period": "7d"}].
*
* Examples:
*
* 1. Allow users to turn on/off pii masking
*
* ```bash
* curl --location 'http://0.0.0.0:4000/key/generate' --header "Authorization: Bearer $LITELLM_MASTER_KEY" --header 'Content-Type: application/json' --data '{
* "permissions": {"allow_pii_controls": true}
* }'
* ```
*
* Returns:
* - key: (str) The generated api key
* - expires: (datetime) Datetime object for when key expires.
* - user_id: (str) Unique user id - used for tracking spend across multiple keys for same user id.
*/
post: operations["generate_key_fn_key_generate_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/key/health": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Key Health
* @description Check the health of the key
*
* Checks:
* - If key based logging is configured correctly - sends a test log
*
* Usage
*
* Pass the key in the request header
*
* ```bash
* curl -X POST "http://localhost:4000/key/health" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -H "Content-Type: application/json"
* ```
*
* Response when logging callbacks are setup correctly:
*
* ```json
* {
* "key": "healthy",
* "logging_callbacks": {
* "callbacks": [
* "gcs_bucket"
* ],
* "status": "healthy",
* "details": "No logger exceptions triggered, system is healthy. Manually check if logs were sent to ['gcs_bucket']"
* }
* }
* ```
*
*
* Response when logging callbacks are not setup correctly:
* ```json
* {
* "key": "unhealthy",
* "logging_callbacks": {
* "callbacks": [
* "gcs_bucket"
* ],
* "status": "unhealthy",
* "details": "Logger exceptions triggered, system is unhealthy: Failed to load vertex credentials. Check to see if credentials containing partial/invalid information."
* }
* }
* ```
*/
post: operations["key_health_key_health_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/key/info": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Info Key Fn
* @description Retrieve information about a key.
*
* Parameters:
* - key: str | None (query parameter) - The key to look up. Accepts the plaintext key or its hash;
* prefer the hash, since a query parameter is recorded verbatim by any HTTP access log in front
* of the proxy. Defaults to the key in the Authorization header.
*
* Returns:
* - key: str - The key that was looked up, echoed back as it was passed in
* - info: dict - The key's row, minus the hashed token. Deleted keys are served from the
* LiteLLM_DeletedVerificationToken archive and carry deleted_at / deleted_by
* - status: "active" | "expired" | "revoked" | "deleted" - Derived from blocked, expires and
* whether the row came from the archive
* - key_alias: str | None - User-friendly key alias
* - spend: float - Amount spent by the key. When budget_duration is set this covers only the
* current budget window, not the key's lifetime
* - max_budget: float | None - Max budget for the key, enforced against spend
* - budget_duration: str | None - Budget reset period ("30d", "1h", etc.)
* - budget_reset_at: datetime | None - When the current budget window ends and spend is next
* reset to 0, not when it was last reset. Reset times snap to standard boundaries in the
* configured timezone (30d and 1mo land on the 1st of the month, 7d on Monday, 1h on the
* hour), so subtracting budget_duration from it does not give the window's start
* - model_max_budget: dict - Per-model budgets, e.g. {"gpt-4": {"budget_limit": 0.0005, "time_period": "30d"}}
* - model_max_budget_usage: dict | None - Current-window spend per model, present only when
* the key has per-model budgets. Each entry's budget_reset_at is when that model's open
* window ends and its spend resets, or None when no window is open or it ends after year 9999
* - budget_limits: list | None - Concurrent budget windows, exactly as stored
* - budget_limits_usage: dict | None - Current-window spend per budget window, e.g.
* {"1h": {"current_spend": 0.0009}}, present only when the key has budget windows
* (read from the same cross-pod spend counter the budget enforcement uses)
* - models: list - Model_name's the key is allowed to call
* - tpm_limit / rpm_limit: int | None - Tokens and requests per minute limits
* - metadata: dict - Metadata for the key, e.g. {"team": "core-infra"}
* - blocked: bool | None - Whether the key is blocked
* - expires: datetime | None - When the key stops authenticating requests
* - last_active: datetime | None - When the key was last used
* - object_permission: dict | None - Resolved vector store / MCP permissions when the key has
* an object_permission_id
*
* Example Curl:
* ```
* curl -X GET "http://0.0.0.0:4000/key/info?key=d5345c0ecc68ae6295c69f91926b2bd379e25481a40c34b5884d157a9f65d8fa" -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*
* Example Curl - if no key is passed, it will use the Key Passed in Authorization Header
* ```
* curl -X GET "http://0.0.0.0:4000/key/info" -H "Authorization: Bearer sk-test-example-key-123"
* ```
*/
get: operations["info_key_fn_key_info_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/key/list": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List Keys
* @description List all keys for a given user / team / organization.
*
* Parameters:
* expand: Optional[List[str]] - Expand related objects (e.g. 'user' to include user information)
* status: Optional[str] - Filter by status: "active", "expired", "revoked" (blocked) or "deleted".
* "deleted" reads the LiteLLM_DeletedVerificationToken archive; the other values partition the
* live key table, so every live key matches exactly one of them.
*
* Returns:
* {
* "keys": List[str] or List[UserAPIKeyAuth],
* "total_count": int,
* "current_page": int,
* "total_pages": int,
* }
*
* When expand includes "user", each key object will include a "user" field with the associated user object.
* Note: When expand=user is specified, full key objects are returned regardless of the return_full_object parameter.
*/
get: operations["list_keys_key_list_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/key/regenerate": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Regenerate Key Fn
* @description Regenerate an existing API key while optionally updating its parameters.
*
* Parameters:
* - key: str (path parameter) - The key to regenerate
* - data: Optional[RegenerateKeyRequest] - Request body containing optional parameters to update
* - key: Optional[str] - The key to regenerate.
* - new_master_key: Optional[str] - The new master key to use, if key is the master key.
* - new_key: Optional[str] - The new key to use, if key is not the master key. Must start with 'sk-' and be at least 16 characters long. If both set, new_master_key will be used.
* - key_alias: Optional[str] - User-friendly key alias
* - user_id: Optional[str] - User ID associated with key
* - team_id: Optional[str] - Team ID associated with key
* - models: Optional[list] - Model_name's a user is allowed to call
* - tags: Optional[List[str]] - Tags for organizing keys (Enterprise only)
* - spend: Optional[float] - Amount spent by key
* - max_budget: Optional[float] - Max budget for key
* - model_max_budget: Optional[Dict[str, BudgetConfig]] - Model-specific budgets {"gpt-4": {"budget_limit": 0.0005, "time_period": "30d"}}
* - budget_fallbacks: Optional[Dict[str, List[str]]] - Per-model fallback chain tried in order when that model's own `model_max_budget` is exceeded, e.g. {"gpt-4o": ["gpt-4o-mini"]}.
* - budget_duration: Optional[str] - Budget reset period ("30d", "1h", etc.)
* - soft_budget: Optional[float] - Soft budget limit (warning vs. hard stop). Will trigger a slack alert when this soft budget is reached.
* - max_parallel_requests: Optional[int] - Rate limit for parallel requests
* - metadata: Optional[dict] - Metadata for key. Example {"team": "core-infra", "app": "app2"}
* - tpm_limit: Optional[int] - Tokens per minute limit
* - rpm_limit: Optional[int] - Requests per minute limit
* - model_rpm_limit: Optional[dict] - Model-specific RPM limits {"gpt-4": 100, "claude-v1": 200}
* - model_tpm_limit: Optional[dict] - Model-specific TPM limits {"gpt-4": 100000, "claude-v1": 200000}
* - allowed_cache_controls: Optional[list] - List of allowed cache control values
* - duration: Optional[str] - Key validity duration ("30d", "1h", etc.)
* - permissions: Optional[dict] - Key-specific permissions
* - guardrails: Optional[List[str]] - List of active guardrails for the key
* - blocked: Optional[bool] - Whether the key is blocked
* - grace_period: Optional[str] - Duration to keep old key valid after rotation (e.g. "24h", "2d"). Omitted = immediate revoke. Env: LITELLM_KEY_ROTATION_GRACE_PERIOD
*
*
* Returns:
* - GenerateKeyResponse containing the new key and its updated parameters
*
* Example:
* ```bash
* curl --location --request POST "http://localhost:4000/key/$LITELLM_API_KEY/regenerate" --header "Authorization: Bearer $LITELLM_MASTER_KEY" --header 'Content-Type: application/json' --data-raw '{
* "max_budget": 100,
* "metadata": {"team": "core-infra"},
* "models": ["gpt-4", "gpt-3.5-turbo"]
* }'
* ```
*
* Note: This is an Enterprise feature. It requires a premium license to use.
*/
post: operations["regenerate_key_fn_key_regenerate_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/key/service-account/generate": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Generate Service Account Key Fn
* @description Generate a Service Account API key based on the provided data. This key does not belong to any user. It belongs to the team.
*
* Why use a service account key?
* - Prevent key from being deleted when user is deleted.
* - Apply team limits, not team member limits to key.
*
* Docs: https://docs.litellm.ai/docs/proxy/virtual_keys
*
* Parameters:
* - duration: Optional[str] - Specify the length of time the token is valid for. You can set duration as seconds ("30s"), minutes ("30m"), hours ("30h"), days ("30d").
* - key_alias: Optional[str] - User defined key alias
* - key: Optional[str] - User defined key value. Must start with 'sk-' and be at least 16 characters long. If not set, a 16-digit unique sk-key is created for you.
* - team_id: Optional[str] - The team id of the key
* - user_id: Optional[str] - [NON-FUNCTIONAL] THIS WILL BE IGNORED. The user id of the key
* - budget_id: Optional[str] - The budget id associated with the key. Created by calling `/budget/new`.
* - end_user_budget_id: Optional[str] - Proxy admin only. Budget id applied to end users first seen through this key that carry no budget of their own. Omit to keep the current value, pass an empty string to clear it.
* - models: Optional[list] - Model_name's a user is allowed to call. (if empty, key is allowed to call all models)
* - aliases: Optional[dict] - Any alias mappings, on top of anything in the config.yaml model list. - https://docs.litellm.ai/docs/proxy/virtual_keys#managing-auth---upgradedowngrade-models
* - config: Optional[dict] - any key-specific configs, overrides config in config.yaml
* - spend: Optional[int] - Amount spent by key. Default is 0. Will be updated by proxy whenever key is used. https://docs.litellm.ai/docs/proxy/virtual_keys#managing-auth---tracking-spend
* - send_invite_email: Optional[bool] - Whether to send an invite email to the user_id, with the generate key
* - max_budget: Optional[float] - Specify max budget for a given key.
* - budget_duration: Optional[str] - Budget is reset at the end of specified duration. If not set, budget is never reset. You can set duration as seconds ("30s"), minutes ("30m"), hours ("30h"), days ("30d").
* - max_parallel_requests: Optional[int] - Rate limit a user based on the number of parallel requests. Raises 429 error, if user's parallel requests > x.
* - metadata: Optional[dict] - Metadata for key, store information for key. Example metadata = {"team": "core-infra", "app": "app2", "email": "[email protected]" }
* - guardrails: Optional[List[str]] - List of active guardrails for the key
* - permissions: Optional[dict] - key-specific permissions. Currently just used for turning off pii masking (if connected). Example - {"pii": false}
* - model_max_budget: Optional[Dict[str, BudgetConfig]] - Model-specific budgets {"gpt-4": {"budget_limit": 0.0005, "time_period": "30d"}}}. IF null or {} then no model specific budget.
* - budget_fallbacks: Optional[Dict[str, List[str]]] - Per-model fallback chain tried in order when that model's own `model_max_budget` is exceeded, e.g. {"gpt-4o": ["gpt-4o-mini"]}.
* - model_rpm_limit: Optional[dict] - key-specific model rpm limit. Example - {"text-davinci-002": 1000, "gpt-3.5-turbo": 1000}. IF null or {} then no model specific rpm limit.
* - model_tpm_limit: Optional[dict] - key-specific model tpm limit. Example - {"text-davinci-002": 1000, "gpt-3.5-turbo": 1000}. IF null or {} then no model specific tpm limit.
* - default_estimated_output_tokens: Optional[int] - Proxy admin only. Expected output tokens reserved for TPM limiting when a request omits max_tokens. Positive integer. Falls back to the team setting, then to the built-in estimate.
* - default_estimated_output_tokens_per_model: Optional[dict] - Proxy admin only. Per-model override of the above. Example - {"gpt-4": 4096, "gpt-3.5-turbo": 1024}. Takes precedence over the key-wide value.
* - mcp_rpm_limit: Optional[dict] - key-specific per-MCP-server rpm limit, keyed by MCP server name (alias if set, else the configured name). Example - {"github": 100, "slack": 200}. IF null or {} then no MCP-specific rpm limit.
* - tpm_limit_type: Optional[str] - TPM rate limit type - "best_effort_throughput", "guaranteed_throughput", or "dynamic"
* - rpm_limit_type: Optional[str] - RPM rate limit type - "best_effort_throughput", "guaranteed_throughput", or "dynamic"
* - allowed_cache_controls: Optional[list] - List of allowed cache control values. Example - ["no-cache", "no-store"]. See all values - https://docs.litellm.ai/docs/proxy/caching#turn-on--off-caching-per-request
* - blocked: Optional[bool] - Whether the key is blocked.
* - rpm_limit: Optional[int] - Specify rpm limit for a given key (Requests per minute)
* - tpm_limit: Optional[int] - Specify tpm limit for a given key (Tokens per minute)
* - tpd_limit: Optional[int] - Specify tpd limit for a given key (Tokens per day). Charged by batch submissions instead of tpm_limit/rpm_limit.
* - soft_budget: Optional[float] - Specify soft budget for a given key. Will trigger a slack alert when this soft budget is reached.
* - tags: Optional[List[str]] - Tags for [tracking spend](https://litellm.vercel.app/docs/proxy/enterprise#tracking-spend-for-custom-tags) and/or doing [tag-based routing](https://litellm.vercel.app/docs/proxy/tag_routing).
* - enforced_params: Optional[List[str]] - List of enforced params for the key (Enterprise only). [Docs](https://docs.litellm.ai/docs/proxy/enterprise#enforce-required-params-for-llm-requests)
* - allowed_routes: Optional[list] - List of allowed routes for the key. Store the actual route or store a wildcard pattern for a set of routes. Example - ["/chat/completions", "/embeddings", "/keys/*"]
* - object_permission: Optional[LiteLLM_ObjectPermissionBase] - key-specific object permission. Example - {"vector_stores": ["vector_store_1", "vector_store_2"], "agents": ["agent_1", "agent_2"], "agent_access_groups": ["dev_group"]}. IF null or {} then no object permission.
* Examples:
* - allowed_vector_store_indexes: Optional[List[dict]] - List of allowed vector store indexes for the key. Example - [{"index_name": "my-index", "index_permissions": ["write", "read"]}]. If specified, the key will only be able to use these specific vector store indexes. Create index, using `/v1/indexes` endpoint.
*
*
* 1. Allow users to turn on/off pii masking
*
* ```bash
* curl --location 'http://0.0.0.0:4000/key/generate' --header "Authorization: Bearer $LITELLM_MASTER_KEY" --header 'Content-Type: application/json' --data '{
* "permissions": {"allow_pii_controls": true}
* }'
* ```
*
* Returns:
* - key: (str) The generated api key
* - expires: (datetime) Datetime object for when key expires.
* - user_id: (str) Unique user id - used for tracking spend across multiple keys for same user id.
*/
post: operations["generate_service_account_key_fn_key_service_account_generate_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/key/spend/report": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Key Spend Report
* @description Get spend for the calling api_key over a date range, with a per-model breakdown.
*
* Same row shape as `/global/spend/report?api_key=...`, but callable by any key:
* non-admin callers are always scoped to their own api_key, while proxy admins
* may pass `?api_key=` to view any key.
*/
get: operations["get_key_spend_report_key_spend_report_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/key/unblock": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Unblock Key
* @description Unblock a Virtual key to allow it to make requests again.
*
* Parameters:
* - key: str - The key to unblock. Can be either the unhashed key (sk-...) or the hashed key value
*
* Example:
* ```bash
* curl --location 'http://0.0.0.0:4000/key/unblock' --header "Authorization: Bearer $LITELLM_MASTER_KEY" --header 'Content-Type: application/json' --data '{
* "key": "sk-Fn8Ej39NxjAXrvpUGKghGw"
* }'
* ```
*
* Note: This is an admin-only endpoint. Only proxy admins, team admins, or org admins can unblock keys.
*/
post: operations["unblock_key_key_unblock_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/key/update": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Update Key Fn
* @description Update an existing API key's parameters.
*
* The body is a merge patch: a field left out keeps its stored value, and on the key's own columns
* an explicit null clears it. The metadata-backed fields below are the exception, merging into the
* stored metadata instead: passing one as null leaves it unchanged, while `metadata` itself
* replaces the stored metadata wholesale.
*
* Parameters:
* - key: Optional[str] - The key to update. Either key or key_alias must be provided.
* - key_alias: Optional[str] - User-friendly key alias. If key is omitted, also identifies the key to update (must match exactly one key, same as /key/delete's key_aliases)
* - user_id: Optional[str] - User ID associated with key
* - team_id: Optional[str] - Team ID associated with key
* - agent_id: Optional[str] - The agent id associated with the key.
* - project_id: Optional[str] - Omit to retain the project, or send null to detach. A different project ID is rejected.
* - organization_id: Optional[str] - The organization id of the key.
* - budget_id: Optional[str] - The budget id associated with the key. Created by calling `/budget/new`.
* - end_user_budget_id: Optional[str] - Proxy admin only. Budget id applied to end users first seen through this key that carry no budget of their own. Omit to keep the current value, pass an empty string to clear it.
* - models: Optional[list] - Model_name's a user is allowed to call
* - tags: Optional[List[str]] - Tags for organizing keys (Enterprise only)
* - prompts: Optional[List[str]] - List of prompts that the key is allowed to use.
* - enforced_params: Optional[List[str]] - List of enforced params for the key (Enterprise only). [Docs](https://docs.litellm.ai/docs/proxy/enterprise#enforce-required-params-for-llm-requests)
* - spend: Optional[float] - Amount spent by key
* - max_budget: Optional[float] - Max budget for key
* - model_max_budget: Optional[Dict[str, BudgetConfig]] - Model-specific budgets {"gpt-4": {"budget_limit": 0.0005, "time_period": "30d"}}
* - budget_fallbacks: Optional[Dict[str, List[str]]] - Per-model fallback chain tried in order when that model's own `model_max_budget` is exceeded, e.g. {"gpt-4o": ["gpt-4o-mini"]}.
* - budget_duration: Optional[str] - Budget reset period ("30d", "1h", etc.)
* - soft_budget: Optional[float] - Soft budget limit (warning vs. hard stop). Will trigger a slack alert when this soft budget is reached. Set to null to remove the soft budget.
* - max_parallel_requests: Optional[int] - Rate limit for parallel requests
* - metadata: Optional[dict] - Metadata for key. Example {"team": "core-infra", "app": "app2"}
* - tpm_limit: Optional[int] - Tokens per minute limit
* - rpm_limit: Optional[int] - Requests per minute limit
* - tpd_limit: Optional[int] - Tokens per day limit, charged by batch submissions instead of tpm_limit/rpm_limit
* - model_rpm_limit: Optional[dict] - Model-specific RPM limits {"gpt-4": 100, "claude-v1": 200}
* - mcp_rpm_limit: Optional[dict] - Per-MCP-server RPM limits, keyed by MCP server name {"github": 100, "slack": 200}
* - tag_rpm_limit: Optional[dict] - Per-request-tag RPM limits, keyed by request tag {"cell-1": 1000, "cell-2": 500}. Each tag gets an independent counter; absent tags fall back to the key-level rpm limit.
* - model_tpm_limit: Optional[dict] - Model-specific TPM limits {"gpt-4": 100000, "claude-v1": 200000}
* - default_estimated_output_tokens: Optional[int] - Proxy admin only. Expected output tokens reserved for TPM limiting when a request omits max_tokens. Positive integer.
* - default_estimated_output_tokens_per_model: Optional[dict] - Proxy admin only. Per-model override of the above {"gpt-4": 4096, "gpt-3.5-turbo": 1024}
* - tpm_limit_type: Optional[str] - TPM rate limit type - "best_effort_throughput", "guaranteed_throughput", or "dynamic"
* - rpm_limit_type: Optional[str] - RPM rate limit type - "best_effort_throughput", "guaranteed_throughput", or "dynamic"
* - allowed_cache_controls: Optional[list] - List of allowed cache control values
* - duration: Optional[str] - Key validity duration ("30d", "1h", etc.), null to never expire, or "-1" to never expire (deprecated, use null)
* - permissions: Optional[dict] - Key-specific permissions
* - send_invite_email: Optional[bool] - Send invite email to user_id
* - guardrails: Optional[List[str]] - List of active guardrails for the key
* - policies: Optional[List[str]] - List of policy names to apply to the key. Policies define guardrails, conditions, and inheritance rules.
* - disable_global_guardrails: Optional[bool] - Whether to disable global guardrails for the key. Proxy admin only.
* - throttle_on_budget_exceeded: Optional[bool] - When the key exceeds its max_budget, throttle its tpm/rpm to the global budget_exceeded_throttle_percentage instead of blocking the key entirely.
* - enable_prompt_caching: Optional[bool] - Auto-inject prompt caching breakpoints (Anthropic cache_control markers) on requests made with this key. Supported Claude models on Anthropic, Bedrock, Vertex AI, and Azure AI only.
* - prompts: Optional[List[str]] - List of prompts that the key is allowed to use.
* - blocked: Optional[bool] - Whether the key is blocked
* - aliases: Optional[dict] - Model aliases for the key - [Docs](https://litellm.vercel.app/docs/proxy/virtual_keys#model-aliases)
* - config: Optional[dict] - [DEPRECATED PARAM] Key-specific config.
* - temp_budget_increase: Optional[float] - Temporary budget increase for the key (Enterprise only).
* - temp_budget_expiry: Optional[str] - Expiry time for the temporary budget increase (Enterprise only).
* - allowed_routes: Optional[list] - List of allowed routes for the key. Store the actual route or store a wildcard pattern for a set of routes. Example - ["/chat/completions", "/embeddings", "/keys/*"]
* - allowed_passthrough_routes: Optional[list] - List of allowed pass through routes for the key. Store the actual route or store a wildcard pattern for a set of routes. Example - ["/my-custom-endpoint"]. Use this instead of allowed_routes, if you just want to specify which pass through routes the key can access, without specifying the routes. If allowed_routes is specified, allowed_passthrough_routes is ignored.
* - denied_passthrough_routes: Optional[list] - List of pass through routes the key may not call, even if allowed by `allowed_passthrough_routes` or `allowed_routes`. Matches exact paths, path prefixes, and trailing `*` wildcards. Applies together with the team's `denied_passthrough_routes`. Example - ["/my-custom-endpoint/admin"].
* - prompts: Optional[List[str]] - List of allowed prompts for the key. If specified, the key will only be able to use these specific prompts.
* - object_permission: Optional[LiteLLM_ObjectPermissionBase] - key-specific object permission. Example - {"vector_stores": ["vector_store_1", "vector_store_2"], "agents": ["agent_1", "agent_2"], "agent_access_groups": ["dev_group"]}. IF null or {} then no object permission.
* - auto_rotate: Optional[bool] - Whether this key should be automatically rotated
* - rotation_interval: Optional[str] - How often to rotate this key (e.g., '30d', '90d'). Required if auto_rotate=True
* - allowed_vector_store_indexes: Optional[List[dict]] - List of allowed vector store indexes for the key. Example - [{"index_name": "my-index", "index_permissions": ["write", "read"]}]. If specified, the key will only be able to use these specific vector store indexes. Create index, using `/v1/indexes` endpoint.
* - router_settings: Optional[UpdateRouterConfig] - key-specific router settings. Example - {"model_group_retry_policy": {"gpt-4": {"RateLimitErrorRetries": 5}}}. IF null or {} then no router settings.
* - access_group_ids: Optional[List[str]] - List of access group IDs to associate with the key. Access groups define which models a key can access. Example - ["access_group_1", "access_group_2"].
* - budget_limits: Optional[list] - List of concurrent budget windows for the key. Each window specifies a budget_limit, time_period, and optional budget_duration. Example - [{"budget_limit": 10.0, "time_period": "1d"}, {"budget_limit": 50.0, "time_period": "7d"}].
*
* Example:
* ```bash
* curl --location 'http://0.0.0.0:4000/key/update' --header "Authorization: Bearer $LITELLM_MASTER_KEY" --header 'Content-Type: application/json' --data '{
* "key": "sk-<your-virtual-key>",
* "key_alias": "my-key",
* "user_id": "user-1234",
* "team_id": "team-1234",
* "max_budget": 100,
* "metadata": {"any_key": "any-val"},
* }'
* ```
*/
post: operations["update_key_fn_key_update_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/key/{key}/regenerate": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Regenerate Key Fn
* @description Regenerate an existing API key while optionally updating its parameters.
*
* Parameters:
* - key: str (path parameter) - The key to regenerate
* - data: Optional[RegenerateKeyRequest] - Request body containing optional parameters to update
* - key: Optional[str] - The key to regenerate.
* - new_master_key: Optional[str] - The new master key to use, if key is the master key.
* - new_key: Optional[str] - The new key to use, if key is not the master key. Must start with 'sk-' and be at least 16 characters long. If both set, new_master_key will be used.
* - key_alias: Optional[str] - User-friendly key alias
* - user_id: Optional[str] - User ID associated with key
* - team_id: Optional[str] - Team ID associated with key
* - models: Optional[list] - Model_name's a user is allowed to call
* - tags: Optional[List[str]] - Tags for organizing keys (Enterprise only)
* - spend: Optional[float] - Amount spent by key
* - max_budget: Optional[float] - Max budget for key
* - model_max_budget: Optional[Dict[str, BudgetConfig]] - Model-specific budgets {"gpt-4": {"budget_limit": 0.0005, "time_period": "30d"}}
* - budget_fallbacks: Optional[Dict[str, List[str]]] - Per-model fallback chain tried in order when that model's own `model_max_budget` is exceeded, e.g. {"gpt-4o": ["gpt-4o-mini"]}.
* - budget_duration: Optional[str] - Budget reset period ("30d", "1h", etc.)
* - soft_budget: Optional[float] - Soft budget limit (warning vs. hard stop). Will trigger a slack alert when this soft budget is reached.
* - max_parallel_requests: Optional[int] - Rate limit for parallel requests
* - metadata: Optional[dict] - Metadata for key. Example {"team": "core-infra", "app": "app2"}
* - tpm_limit: Optional[int] - Tokens per minute limit
* - rpm_limit: Optional[int] - Requests per minute limit
* - model_rpm_limit: Optional[dict] - Model-specific RPM limits {"gpt-4": 100, "claude-v1": 200}
* - model_tpm_limit: Optional[dict] - Model-specific TPM limits {"gpt-4": 100000, "claude-v1": 200000}
* - allowed_cache_controls: Optional[list] - List of allowed cache control values
* - duration: Optional[str] - Key validity duration ("30d", "1h", etc.)
* - permissions: Optional[dict] - Key-specific permissions
* - guardrails: Optional[List[str]] - List of active guardrails for the key
* - blocked: Optional[bool] - Whether the key is blocked
* - grace_period: Optional[str] - Duration to keep old key valid after rotation (e.g. "24h", "2d"). Omitted = immediate revoke. Env: LITELLM_KEY_ROTATION_GRACE_PERIOD
*
*
* Returns:
* - GenerateKeyResponse containing the new key and its updated parameters
*
* Example:
* ```bash
* curl --location --request POST "http://localhost:4000/key/$LITELLM_API_KEY/regenerate" --header "Authorization: Bearer $LITELLM_MASTER_KEY" --header 'Content-Type: application/json' --data-raw '{
* "max_budget": 100,
* "metadata": {"team": "core-infra"},
* "models": ["gpt-4", "gpt-3.5-turbo"]
* }'
* ```
*
* Note: This is an Enterprise feature. It requires a premium license to use.
*/
post: operations["regenerate_key_fn_key__key__regenerate_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/key/{key}/reset_spend": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/** Reset Key Spend Fn */
post: operations["reset_key_spend_fn_key__key__reset_spend_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/langfuse/{endpoint}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Langfuse Proxy Route
* @description Call Langfuse via LiteLLM proxy. Works with Langfuse SDK.
*
* [Docs](https://docs.litellm.ai/docs/pass_through/langfuse)
*/
get: operations["langfuse_proxy_route_langfuse__endpoint__get"];
/**
* Langfuse Proxy Route
* @description Call Langfuse via LiteLLM proxy. Works with Langfuse SDK.
*
* [Docs](https://docs.litellm.ai/docs/pass_through/langfuse)
*/
put: operations["langfuse_proxy_route_langfuse__endpoint__put"];
/**
* Langfuse Proxy Route
* @description Call Langfuse via LiteLLM proxy. Works with Langfuse SDK.
*
* [Docs](https://docs.litellm.ai/docs/pass_through/langfuse)
*/
post: operations["langfuse_proxy_route_langfuse__endpoint__post"];
/**
* Langfuse Proxy Route
* @description Call Langfuse via LiteLLM proxy. Works with Langfuse SDK.
*
* [Docs](https://docs.litellm.ai/docs/pass_through/langfuse)
*/
delete: operations["langfuse_proxy_route_langfuse__endpoint__delete"];
options?: never;
head?: never;
/**
* Langfuse Proxy Route
* @description Call Langfuse via LiteLLM proxy. Works with Langfuse SDK.
*
* [Docs](https://docs.litellm.ai/docs/pass_through/langfuse)
*/
patch: operations["langfuse_proxy_route_langfuse__endpoint__patch"];
trace?: never;
};
"/laya/v1/systemone": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/** Laya Proxy Route */
post: operations["laya_proxy_route_laya_v1_systemone_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/lazy/warm/{name}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/** Warm */
post: operations["warm_lazy_warm__name__post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/lens": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Lens Request */
get: operations["lens_request_lens_get"];
/** Lens Request */
put: operations["lens_request_lens_put"];
/** Lens Request */
post: operations["lens_request_lens_post"];
/** Lens Request */
delete: operations["lens_request_lens_delete"];
options?: never;
head?: never;
/** Lens Request */
patch: operations["lens_request_lens_patch"];
trace?: never;
};
"/lens/service": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Service Connection */
get: operations["service_connection_lens_service_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/lens/{path}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Lens Request */
get: operations["lens_request_lens__path__get"];
/** Lens Request */
put: operations["lens_request_lens__path__put"];
/** Lens Request */
post: operations["lens_request_lens__path__post"];
/** Lens Request */
delete: operations["lens_request_lens__path__delete"];
options?: never;
head?: never;
/** Lens Request */
patch: operations["lens_request_lens__path__patch"];
trace?: never;
};
"/liteadmin/slack/connect/{token}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Connect Page */
get: operations["connect_page_liteadmin_slack_connect__token__get"];
put?: never;
/** Connect Account */
post: operations["connect_account_liteadmin_slack_connect__token__post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/litellm/.well-known/litellm-ui-config": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get Ui Config */
get: operations["get_ui_config_litellm__well_known_litellm_ui_config_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/live/sessions": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* WebSocket: live_sessions_websocket_endpoint
* @description WebSocket connection endpoint
*/
get: operations["websocket_live_sessions_websocket_endpoint_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/login": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/** Login */
post: operations["login_login_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/management/v1/budgets": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List Budgets
* @description The budgets defined on this proxy, paged, sortable and filterable, for the
* Budgets page.
*
* Readable by a proxy admin or an admin viewer; anyone else is refused 403. The
* older `/budget/list` answers with the whole table as a bare array and has no
* way to page, sort or filter it.
*
* `sort` takes a comma-separated list of `budget_id`, `max_budget`, `tpm_limit`,
* `rpm_limit`, `tpd_limit` or `created_at`, each optionally prefixed with `-` for descending,
* and defaults to `-created_at`. `budget_id` is appended to every sort as the
* tiebreaker. `q` is a case-insensitive substring match on `budget_id`.
* `page_size` defaults to 50 and is capped at 100. Filters are
* `filter[budget_duration][in|is_null]`, `filter[max_budget][gte|lte|is_null]`
* and `filter[created_at][gte|lte]`.
*
* Example curl:
* ```
* curl --location --globoff 'http://0.0.0.0:4000/management/v1/budgets?sort=-max_budget&filter[budget_duration][in]=7d,30d&page_size=25' --header "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*/
get: operations["list_budgets_management_v1_budgets_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/management/v1/spend_logs/end_users": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List Spend Log End Users
* @description The distinct end users appearing in spend logs over a time window, for the logs
* page filter dropdown.
*
* Scoped like `/spend/logs/ui`: a proxy admin sees every end user in the window,
* anyone else sees only end users from their own requests or from teams they
* administer (or hold the `/spend/logs` permission on).
*
* The window is required and the inner scan is capped at SPEND_LOGS_FACET_SCAN_CAP
* rows, so the query cannot degrade into a full-table scan the way
* `/global/all_end_users` does.
*
* Example curl:
* ```
* curl --location --globoff 'http://0.0.0.0:4000/management/v1/spend_logs/end_users?filter[startTime][gte]=2026-07-23T00:00:00Z&filter[startTime][lte]=2026-07-24T00:00:00Z&page_size=50&q=acme' --header "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*/
get: operations["list_spend_log_end_users_management_v1_spend_logs_end_users_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/management/v1/spend_logs/users": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List Spend Log Users
* @description The distinct internal users appearing in spend logs the caller can read.
*/
get: operations["list_spend_log_users_management_v1_spend_logs_users_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/management/v1/teams/{team_id}/members/bulk_delete": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Bulk Delete Team Members Action
* @description Remove up to 500 members from one team in one call. Same authorization as
* `/team/member_delete`: proxy admins, the team's admins, and admins of the team's
* organization. Each member is named by exactly one of `user_id` or `user_email`;
* unknown body fields are a 422 and an unknown team is a 404.
*
* `data` holds one result per requested member, in request order. A row is
* `success: false` with an `error` when it names nobody on the team or repeats an
* earlier row. The roster is rewritten once, under the team's advisory lock, so a
* concurrent member_add is never overwritten from a stale read.
*
* Example curl:
* ```
* curl --location 'http://0.0.0.0:4000/management/v1/teams/team-1/members/bulk_delete' --header "Authorization: Bearer $LITELLM_MASTER_KEY" --header 'Content-Type: application/json' --data '{"members": [{"user_id": "user-1"}, {"user_email": "user-2@example.com"}]}'
* ```
*/
post: operations["bulk_delete_team_members_action_management_v1_teams__team_id__members_bulk_delete_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/management/v1/teams/{team_id}/members/bulk_update": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Bulk Update Team Member Budgets Action
* @description Set per-member limits for up to 500 members of one team in one call. Same
* authorization and member addressing as `/team/member_update`: proxy admins, the team's
* admins, and admins of the team's organization, with each member named by exactly one of
* `user_id` or `user_email`. Unknown body fields are a 422 and an unknown team is a 404.
*
* Each row is a merge patch of that member's limits: a field left out is untouched, a
* field sent as null is cleared, and clearing the last limit drops the member back to the
* team default. A budget row shared by several memberships, the team default included, is
* copied for the member being patched rather than written in place, so one member's new
* cap never lands on anybody else.
*
* `data` holds one result per requested member, in request order, carrying the limits in
* force after the write. A row is `success: false` with an `error` when it names nobody on
* the team or repeats an earlier row. Roles are not part of this route; `/team/member_update`
* still owns them.
*
* Example curl:
* ```
* curl --location 'http://0.0.0.0:4000/management/v1/teams/team-1/members/bulk_update' --header "Authorization: Bearer $LITELLM_MASTER_KEY" --header 'Content-Type: application/json' --data '{"members": [{"user_id": "user-1", "max_budget_in_team": 10}, {"user_email": "user-2@example.com", "max_budget_in_team": 10, "budget_duration": "30d"}]}'
* ```
*/
post: operations["bulk_update_team_member_budgets_action_management_v1_teams__team_id__members_bulk_update_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/management/v1/users/bulk": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Bulk Create Users Route
* @description Create up to 500 internal users in one request, optionally adding each one to teams.
*
* Every entry in `users` takes the same fields as `/user/new`, with two differences: `auto_create_key`
* defaults to `false` (opt in per user to also get a virtual key back) and `send_invite_email` is not
* supported. Unknown fields are rejected with 422. Rows are validated together (duplicate ids or emails,
* unknown teams, roles the caller may not grant), inserted in one statement, and each referenced team is
* written once for all of its new members.
*
* Rows fail independently: a bad row is reported in `data` with `success: false` and an `error`, and the
* other rows still get created. A user that was created but could not be added to one of its teams is
* reported with `success: true`, `teams` listing where they did land, and `error` naming the failed team.
* The whole request is refused with a 403 problem document only if creating the valid rows would exceed
* the license seat limit.
*
* Example curl:
* ```
* curl -X POST "http://localhost:4000/management/v1/users/bulk" \
* -H "Content-Type: application/json" \
* -H "Authorization: Bearer $LITELLM_MASTER_KEY" \
* -d '{
* "users": [
* {"user_email": "a@example.com", "user_role": "internal_user", "teams": ["team-1"]},
* {"user_email": "b@example.com", "user_role": "internal_user", "auto_create_key": true}
* ]
* }'
* ```
*
* Returns `data` (one entry per input row, in order, with `user_id`, `user_email`, `success`, `teams`,
* `key`, `error`) and `meta` with `total_requested`, `created` and `failed`.
*/
post: operations["bulk_create_users_route_management_v1_users_bulk_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/management/v1/users/bulk_delete": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Bulk Delete Users Action
* @description Delete up to 500 users in one call, taking each out of every team it belongs to.
* Same authorization as `/user/delete`: proxy admins may delete anyone, org admins
* only users inside organizations they administer. Unknown body fields are a 422.
*
* `data` holds one result per requested `user_id`, in request order. A row is
* `success: false` with an `error` when the id is unknown, repeated in the request,
* or outside the caller's scope. Rows that pass those checks are deleted together,
* in one transaction, so either all of them go or none does.
*
* Example curl:
* ```
* curl --location 'http://0.0.0.0:4000/management/v1/users/bulk_delete' --header "Authorization: Bearer $LITELLM_MASTER_KEY" --header 'Content-Type: application/json' --data '{"user_ids": ["user-1", "user-2"]}'
* ```
*/
post: operations["bulk_delete_users_action_management_v1_users_bulk_delete_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/management/v1/users/me": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get Current User */
get: operations["get_current_user_management_v1_users_me_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/mcp": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Aggregate Mcp Route
* @description Serve the aggregate MCP endpoint on the bare ``/mcp`` spelling: the
* ``/mcp`` mount cannot match its bare prefix, and the resulting 307 breaks
* MCP clients behind TLS-terminating proxies.
*/
get: operations["aggregate_mcp_route_mcp_get"];
/**
* Aggregate Mcp Route
* @description Serve the aggregate MCP endpoint on the bare ``/mcp`` spelling: the
* ``/mcp`` mount cannot match its bare prefix, and the resulting 307 breaks
* MCP clients behind TLS-terminating proxies.
*/
put: operations["aggregate_mcp_route_mcp_put"];
/**
* Aggregate Mcp Route
* @description Serve the aggregate MCP endpoint on the bare ``/mcp`` spelling: the
* ``/mcp`` mount cannot match its bare prefix, and the resulting 307 breaks
* MCP clients behind TLS-terminating proxies.
*/
post: operations["aggregate_mcp_route_mcp_post"];
/**
* Aggregate Mcp Route
* @description Serve the aggregate MCP endpoint on the bare ``/mcp`` spelling: the
* ``/mcp`` mount cannot match its bare prefix, and the resulting 307 breaks
* MCP clients behind TLS-terminating proxies.
*/
delete: operations["aggregate_mcp_route_mcp_delete"];
/**
* Aggregate Mcp Route
* @description Serve the aggregate MCP endpoint on the bare ``/mcp`` spelling: the
* ``/mcp`` mount cannot match its bare prefix, and the resulting 307 breaks
* MCP clients behind TLS-terminating proxies.
*/
options: operations["aggregate_mcp_route_mcp_options"];
/**
* Aggregate Mcp Route
* @description Serve the aggregate MCP endpoint on the bare ``/mcp`` spelling: the
* ``/mcp`` mount cannot match its bare prefix, and the resulting 307 breaks
* MCP clients behind TLS-terminating proxies.
*/
head: operations["aggregate_mcp_route_mcp_head"];
/**
* Aggregate Mcp Route
* @description Serve the aggregate MCP endpoint on the bare ``/mcp`` spelling: the
* ``/mcp`` mount cannot match its bare prefix, and the resulting 307 breaks
* MCP clients behind TLS-terminating proxies.
*/
patch: operations["aggregate_mcp_route_mcp_patch"];
trace?: never;
};
"/mcp-rest/test/connection": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Test Connection
* @description Test if we can connect to the provided MCP server before adding it
*/
post: operations["test_connection_mcp_rest_test_connection_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/mcp-rest/test/tools/list": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Test Tools List
* @description Preview tools available from MCP server before adding it
*/
post: operations["test_tools_list_mcp_rest_test_tools_list_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/mcp-rest/tools/call": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Call Tool Rest Api
* @description REST API to call a specific MCP tool with the provided arguments
*/
post: operations["call_tool_rest_api_mcp_rest_tools_call_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/mcp-rest/tools/list": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List Tool Rest Api
* @description List all available tools with information about the server they belong to.
*
* Example response:
* {
* "tools": [
* {
* "name": "create_zap",
* "description": "Create a new zap",
* "inputSchema": "tool_input_schema",
* "mcp_info": {
* "server_name": "zapier",
* "logo_url": "https://www.zapier.com/logo.png",
* "server_id": "a1b2c3d4-...",
* "alias": "zapier_prod",
* }
* }
* ],
* "error": null,
* "message": "Successfully retrieved tools"
* }
*/
get: operations["list_tool_rest_api_mcp_rest_tools_list_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/mcp/proxy": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Proxy Mcp Route
* @description Serve the fixed three-tool MCP proxy surface.
*/
get: operations["proxy_mcp_route_mcp_proxy_get"];
/**
* Proxy Mcp Route
* @description Serve the fixed three-tool MCP proxy surface.
*/
put: operations["proxy_mcp_route_mcp_proxy_put"];
/**
* Proxy Mcp Route
* @description Serve the fixed three-tool MCP proxy surface.
*/
post: operations["proxy_mcp_route_mcp_proxy_post"];
/**
* Proxy Mcp Route
* @description Serve the fixed three-tool MCP proxy surface.
*/
delete: operations["proxy_mcp_route_mcp_proxy_delete"];
/**
* Proxy Mcp Route
* @description Serve the fixed three-tool MCP proxy surface.
*/
options: operations["proxy_mcp_route_mcp_proxy_options"];
/**
* Proxy Mcp Route
* @description Serve the fixed three-tool MCP proxy surface.
*/
head: operations["proxy_mcp_route_mcp_proxy_head"];
/**
* Proxy Mcp Route
* @description Serve the fixed three-tool MCP proxy surface.
*/
patch: operations["proxy_mcp_route_mcp_proxy_patch"];
trace?: never;
};
"/memory-usage-in-mem-cache": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Memory Usage In Mem Cache
* @description 1. user_api_key_cache
* 2. router_cache
* 3. proxy_logging_cache
* 4. internal_usage_cache
*/
get: operations["memory_usage_in_mem_cache_memory_usage_in_mem_cache_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/memory-usage-in-mem-cache-items": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Memory Usage In Mem Cache Items
* @description 1. user_api_key_cache
* 2. router_cache
* 3. proxy_logging_cache
* 4. internal_usage_cache
*/
get: operations["memory_usage_in_mem_cache_items_memory_usage_in_mem_cache_items_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/milvus/{endpoint}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Milvus Proxy Route
* @description Enable using Milvus `/vectors` endpoint as a pass-through endpoint.
*/
get: operations["milvus_proxy_route_milvus__endpoint__get"];
/**
* Milvus Proxy Route
* @description Enable using Milvus `/vectors` endpoint as a pass-through endpoint.
*/
put: operations["milvus_proxy_route_milvus__endpoint__put"];
/**
* Milvus Proxy Route
* @description Enable using Milvus `/vectors` endpoint as a pass-through endpoint.
*/
post: operations["milvus_proxy_route_milvus__endpoint__post"];
/**
* Milvus Proxy Route
* @description Enable using Milvus `/vectors` endpoint as a pass-through endpoint.
*/
delete: operations["milvus_proxy_route_milvus__endpoint__delete"];
options?: never;
head?: never;
/**
* Milvus Proxy Route
* @description Enable using Milvus `/vectors` endpoint as a pass-through endpoint.
*/
patch: operations["milvus_proxy_route_milvus__endpoint__patch"];
trace?: never;
};
"/mistral/{endpoint}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Mistral Proxy Route
* @description [Docs](https://docs.litellm.ai/docs/pass_through/mistral)
*/
get: operations["mistral_proxy_route_mistral__endpoint__get"];
/**
* Mistral Proxy Route
* @description [Docs](https://docs.litellm.ai/docs/pass_through/mistral)
*/
put: operations["mistral_proxy_route_mistral__endpoint__put"];
/**
* Mistral Proxy Route
* @description [Docs](https://docs.litellm.ai/docs/pass_through/mistral)
*/
post: operations["mistral_proxy_route_mistral__endpoint__post"];
/**
* Mistral Proxy Route
* @description [Docs](https://docs.litellm.ai/docs/pass_through/mistral)
*/
delete: operations["mistral_proxy_route_mistral__endpoint__delete"];
options?: never;
head?: never;
/**
* Mistral Proxy Route
* @description [Docs](https://docs.litellm.ai/docs/pass_through/mistral)
*/
patch: operations["mistral_proxy_route_mistral__endpoint__patch"];
trace?: never;
};
"/model-insights": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get Model Insights */
get: operations["get_model_insights_model_insights_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/model-insights/tasks": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get Model Insight Tasks */
get: operations["get_model_insight_tasks_model_insights_tasks_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/model/block": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Block Model
* @description Block a DB-stored model deployment from serving requests.
*
* Parameters:
* - model_id: str - The model deployment id to block.
*/
post: operations["block_model_model_block_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/model/cost_map/source": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Model Cost Map Source
* @description ADMIN ONLY / MASTER KEY Only Endpoint
*
* Returns information about where the current model cost/pricing data was loaded from.
*
* Response fields:
* - source: "local" (bundled backup) or "remote" (fetched from URL)
* - url: the remote URL that was attempted (null when env-forced local)
* - is_env_forced: true if LITELLM_LOCAL_MODEL_COST_MAP=True forced local usage
* - fallback_reason: human-readable reason why remote failed (null on success)
* - loaded_at: when this pod last loaded the map
* - source_revision: git blob id of the loaded file, what git rev-parse <commit>:<path> prints for it
* - etag: the ETag of the remote fetch (null for the bundled backup)
* - model_count: number of models in the currently loaded cost map
*/
get: operations["get_model_cost_map_source_model_cost_map_source_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/model/delete": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Delete Model
* @description Allows deleting models in the model list in the config.yaml
*/
post: operations["delete_model_model_delete_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/model/deprecations": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Model Deprecations
* @description List models with known deprecation/sunset dates, bucketed by urgency.
*
* Reads `deprecation_date` metadata from `model_prices_and_context_window.json`
* (and any per-deployment `model_info.deprecation_date` overrides) for the
* models configured on this proxy. An entry also names its `successor_model`
* when the deployment sets `model_info.successor_model`.
*
* Parameters:
* warn_within_days: Window (in days) used to bucket "imminent" models,
* 30 by default.
*
* Returns:
* A payload with three lists of `ModelDeprecationInfo` entries:
*
* - `deprecated`: deprecation date is in the past, so these requests may
* fail at any time.
* - `imminent`: deprecation date is within `warn_within_days` from today.
* - `upcoming`: deprecation date is further out.
*
* Example:
* ```shell
* curl -X GET 'http://localhost:4000/model/deprecations' \
* -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*/
get: operations["model_deprecations_model_deprecations_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/model/info": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Model Info V1
* @description Provides more info about each model in /models, including config.yaml descriptions (except api key and api base)
*
* Parameters:
* litellm_model_id: Optional[str] = None (this is the value of `x-litellm-model-id` returned in response headers)
*
* - When litellm_model_id is passed, it will return the info for that specific model
* - When litellm_model_id is not passed, it will return the info for all models
* - include_team_models: When true, filter to deployments the caller can use (same as /v2/model/info).
* - teamId: Filter to models accessible by the given team.
* - healthy_only: When true, hide models whose backing deployments are all marked
* unhealthy by background health checks, matching `/v1/models?healthy_only=true`.
* Set `general_settings.model_list_healthy_only: true` to apply this to every
* caller without the query parameter. Requires `background_health_checks: true`,
* plus either `model_list_healthy_only` or `enable_health_check_routing` to keep
* deployment health state cached; without health state the listing is returned
* unfiltered (fail open). Ignored when `litellm_model_id` is passed, since that
* is a direct lookup of one deployment rather than a listing. Hiding is
* presentation-only: a hidden model can still be called directly.
*
* Each model in the list response includes `model_info.access_via_team_ids` and
* `model_info.direct_access` when the proxy database is connected.
*
* Returns:
* A JSON response whose `data` list holds one entry per model.
*
* Example Response:
* ```json
* {
* "data": [
* {
* "model_name": "fake-openai-endpoint",
* "litellm_params": {
* "api_base": "https://exampleopenaiendpoint-production.up.railway.app/",
* "model": "openai/fake"
* },
* "model_info": {
* "id": "112f74fab24a7a5245d2ced3536dd8f5f9192c57ee6e332af0f0512e08bed5af",
* "db_model": false
* }
* }
* ]
* }
*
* ```
*/
get: operations["model_info_v1_model_info_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/model/metrics": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Model Metrics
* @description View number of requests & avg latency per model on config.yaml
*/
get: operations["model_metrics_model_metrics_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/model/metrics/exceptions": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Model Metrics Exceptions
* @description View number of failed requests per model on config.yaml
*/
get: operations["model_metrics_exceptions_model_metrics_exceptions_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/model/metrics/slow_responses": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Model Metrics Slow Responses
* @description View number of hanging requests per model_group
*/
get: operations["model_metrics_slow_responses_model_metrics_slow_responses_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/model/new": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Add New Model
* @description Allows adding new models to the model list in the config.yaml
*/
post: operations["add_new_model_model_new_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/model/settings": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Model Settings
* @description Returns provider name, description, and required parameters for each provider
*/
get: operations["model_settings_model_settings_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/model/streaming_metrics": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Model Streaming Metrics
* @description View time to first token for models in spend logs
*/
get: operations["model_streaming_metrics_model_streaming_metrics_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/model/unblock": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Unblock Model
* @description Unblock a DB-stored model deployment so it can serve requests again.
*
* Parameters:
* - model_id: str - The model deployment id to unblock.
*/
post: operations["unblock_model_model_unblock_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/model/update": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Update Model
* @description Edit existing model params
*/
post: operations["update_model_model_update_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/model/{model_id}/kubernetes_pods": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get Kubernetes Pods */
get: operations["get_kubernetes_pods_model__model_id__kubernetes_pods_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/model/{model_id}/update": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
/**
* Patch Model
* @description PATCH Endpoint for partial model updates.
*
* Only updates the fields specified in the request while preserving other existing values.
* Follows proper PATCH semantics by only modifying provided fields.
*
* Args:
* model_id: The ID of the model to update
* patch_data: The fields to update and their new values
* user_api_key_dict: User authentication information
*
* Returns:
* Updated model information
*
* Raises:
* ProxyException: For various error conditions including authentication and database errors
*/
patch: operations["patch_model_model__model_id__update_patch"];
trace?: never;
};
"/model_group/info": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Model Group Info
* @description Get information about all the deployments on litellm proxy, including config.yaml descriptions (except api key and api base)
*
* - /model_group/info returns all model groups. End users of proxy should use /model_group/info since those models will be used for /chat/completions, /embeddings, etc.
* - /model_group/info?model_group=rerank-english-v3.0 returns all model groups for a specific model group (`model_name` in config.yaml)
*
*
*
* Example Request (All Models):
* ```shell
* curl -X 'GET' 'http://localhost:4000/model_group/info' -H 'accept: application/json' -H "x-api-key: $LITELLM_MASTER_KEY"
* ```
*
* Example Request (Specific Model Group):
* ```shell
* curl -X 'GET' 'http://localhost:4000/model_group/info?model_group=rerank-english-v3.0' -H 'accept: application/json' -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*
* Example Request (Specific Wildcard Model Group): (e.g. `model_name: openai/*` on config.yaml)
* ```shell
* curl -X 'GET' 'http://localhost:4000/model_group/info?model_group=openai/tts-1'
* -H 'accept: application/json' -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*
* Learn how to use and set wildcard models [here](https://docs.litellm.ai/docs/wildcard_routing)
*
* Example Response:
* ```json
* {
* "data": [
* {
* "model_group": "rerank-english-v3.0",
* "providers": [
* "cohere"
* ],
* "max_input_tokens": null,
* "max_output_tokens": null,
* "input_cost_per_token": 0.0,
* "output_cost_per_token": 0.0,
* "mode": null,
* "tpm": null,
* "rpm": null,
* "supports_parallel_function_calling": false,
* "supports_vision": false,
* "supports_function_calling": false,
* "supported_openai_params": [
* "stream",
* "temperature",
* "max_tokens",
* "logit_bias",
* "top_p",
* "frequency_penalty",
* "presence_penalty",
* "stop",
* "n",
* "extra_headers"
* ]
* },
* {
* "model_group": "gpt-3.5-turbo",
* "providers": [
* "openai"
* ],
* "max_input_tokens": 16385.0,
* "max_output_tokens": 4096.0,
* "input_cost_per_token": 1.5e-06,
* "output_cost_per_token": 2e-06,
* "mode": "chat",
* "tpm": null,
* "rpm": null,
* "supports_parallel_function_calling": false,
* "supports_vision": false,
* "supports_function_calling": true,
* "supported_openai_params": [
* "frequency_penalty",
* "logit_bias",
* "logprobs",
* "top_logprobs",
* "max_tokens",
* "max_completion_tokens",
* "n",
* "presence_penalty",
* "seed",
* "stop",
* "stream",
* "stream_options",
* "temperature",
* "top_p",
* "tools",
* "tool_choice",
* "function_call",
* "functions",
* "max_retries",
* "extra_headers",
* "parallel_tool_calls",
* "response_format"
* ]
* },
* {
* "model_group": "llava-hf",
* "providers": [
* "openai"
* ],
* "max_input_tokens": null,
* "max_output_tokens": null,
* "input_cost_per_token": 0.0,
* "output_cost_per_token": 0.0,
* "mode": null,
* "tpm": null,
* "rpm": null,
* "supports_parallel_function_calling": false,
* "supports_vision": true,
* "supports_function_calling": false,
* "supported_openai_params": [
* "frequency_penalty",
* "logit_bias",
* "logprobs",
* "top_logprobs",
* "max_tokens",
* "max_completion_tokens",
* "n",
* "presence_penalty",
* "seed",
* "stop",
* "stream",
* "stream_options",
* "temperature",
* "top_p",
* "tools",
* "tool_choice",
* "function_call",
* "functions",
* "max_retries",
* "extra_headers",
* "parallel_tool_calls",
* "response_format"
* ]
* }
* ]
* }
* ```
*/
get: operations["model_group_info_model_group_info_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/model_group/make_public": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Update Public Model Groups
* @description Update which model groups are public
*/
post: operations["update_public_model_groups_model_group_make_public_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/model_hub/update_useful_links": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Update Useful Links
* @description Update useful links
*/
post: operations["update_useful_links_model_hub_update_useful_links_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/models": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Model List
* @description Use `/model/info` - to get detailed model information, example - pricing, mode, etc.
*
* This is just for compatibility with openai projects like aider.
*
* Query Parameters:
* - client_version: Sent by Codex CLI (`?client_version=0.159.3`) when it fetches a
* provider's model catalog. When present, the response is Codex's own
* catalog shape (`{"models": [...]}`) built from the same listing, with
* each model's `model_info.service_tiers` as its service tiers; absent,
* the OpenAI shape below is returned
* - include_metadata: Include additional metadata in the response with fallback information
* - fallback_type: Type of fallbacks to include ("general", "context_window", "content_policy")
* Defaults to "general" when include_metadata=true
* - scope: Optional scope parameter. Currently only accepts "expand".
* When scope=expand is passed, proxy admins, team admins, and org admins
* will receive all proxy models as if they are a proxy admin.
* - healthy_only: When true, hide models whose backing deployments are all marked
* unhealthy by background health checks. Set
* `general_settings.model_list_healthy_only: true` to apply this
* to every caller without the query parameter. Requires
* `background_health_checks: true` in general_settings, plus
* either `model_list_healthy_only` or `enable_health_check_routing`
* to keep deployment health state cached; without health state
* the listing is returned unfiltered (fail open).
* Models expanded from wildcard routes (e.g. `openai/*`) are not
* filtered, and nothing is hidden when `allowed_fails_policy` is
* configured (cooldown remains the sole exclusion mechanism).
* Hiding is presentation-only: a hidden model can still be
* called directly.
*/
get: operations["model_list_models_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/models/{model_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Model Info
* @description Retrieve information about a specific model accessible to your API key.
*
* Returns model details only if the model is available to your API key/team.
* Returns 404 if the model doesn't exist or is not accessible.
*
* Follows OpenAI API specification for individual model retrieval.
* https://platform.openai.com/docs/api-reference/models/retrieve
*
* Query parameters mirror `/v1/models` so the same caller context (team
* scoping, health filtering, paused deployments) drives both endpoints; the
* listing's public id must resolve to the same internal deployment here.
*/
get: operations["model_info_models__model_id__get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/models/{model_name}:countTokens": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Google Count Tokens
* @description ```json
* return {
* "totalTokens": 31,
* "totalBillableCharacters": 96,
* "promptTokensDetails": [
* {
* "modality": "TEXT",
* "tokenCount": 31
* }
* ]
* }
* ```
*/
post: operations["google_count_tokens_models__model_name__countTokens_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/models/{model_name}:generateContent": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/** Google Generate Content */
post: operations["google_generate_content_models__model_name__generateContent_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/models/{model_name}:streamGenerateContent": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/** Google Stream Generate Content */
post: operations["google_stream_generate_content_models__model_name__streamGenerateContent_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/moderations": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Moderations
* @description The moderations endpoint is a tool you can use to check whether content complies with an LLM Providers policies.
* Quick Start
* ```
* curl --location 'http://0.0.0.0:4000/moderations' --header 'Content-Type: application/json' --header "Authorization: Bearer $LITELLM_MASTER_KEY" --data '{"input": "Sample text goes here", "model": "text-moderation-stable"}'
* ```
*/
post: operations["moderations_moderations_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/moyai/connect/exchange": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/** Moyai Connect Exchange */
post: operations["moyai_connect_exchange_moyai_connect_exchange_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/moyai/connect/start": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/** Moyai Connect Start */
post: operations["moyai_connect_start_moyai_connect_start_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/nvidia_nim/{endpoint}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Nvidia Nim Proxy Route
* @description Relay a native NVIDIA NIM request through a LiteLLM model group.
*
* `{PROXY_BASE_URL}/nvidia_nim/{model_group}/v1/infer` forwards the body unchanged to the deployment's
* `api_base`, so object detection and OCR NIMs whose payload carries no `model` field still go through
* virtual key auth, model access checks, and spend logging.
*/
get: operations["nvidia_nim_proxy_route_nvidia_nim__endpoint__get"];
/**
* Nvidia Nim Proxy Route
* @description Relay a native NVIDIA NIM request through a LiteLLM model group.
*
* `{PROXY_BASE_URL}/nvidia_nim/{model_group}/v1/infer` forwards the body unchanged to the deployment's
* `api_base`, so object detection and OCR NIMs whose payload carries no `model` field still go through
* virtual key auth, model access checks, and spend logging.
*/
put: operations["nvidia_nim_proxy_route_nvidia_nim__endpoint__put"];
/**
* Nvidia Nim Proxy Route
* @description Relay a native NVIDIA NIM request through a LiteLLM model group.
*
* `{PROXY_BASE_URL}/nvidia_nim/{model_group}/v1/infer` forwards the body unchanged to the deployment's
* `api_base`, so object detection and OCR NIMs whose payload carries no `model` field still go through
* virtual key auth, model access checks, and spend logging.
*/
post: operations["nvidia_nim_proxy_route_nvidia_nim__endpoint__post"];
/**
* Nvidia Nim Proxy Route
* @description Relay a native NVIDIA NIM request through a LiteLLM model group.
*
* `{PROXY_BASE_URL}/nvidia_nim/{model_group}/v1/infer` forwards the body unchanged to the deployment's
* `api_base`, so object detection and OCR NIMs whose payload carries no `model` field still go through
* virtual key auth, model access checks, and spend logging.
*/
delete: operations["nvidia_nim_proxy_route_nvidia_nim__endpoint__delete"];
options?: never;
head?: never;
/**
* Nvidia Nim Proxy Route
* @description Relay a native NVIDIA NIM request through a LiteLLM model group.
*
* `{PROXY_BASE_URL}/nvidia_nim/{model_group}/v1/infer` forwards the body unchanged to the deployment's
* `api_base`, so object detection and OCR NIMs whose payload carries no `model` field still go through
* virtual key auth, model access checks, and spend logging.
*/
patch: operations["nvidia_nim_proxy_route_nvidia_nim__endpoint__patch"];
trace?: never;
};
"/ocr": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Ocr
* @description OCR endpoint for extracting text from documents and images.
*
* Supports two input modes:
*
* **1. JSON body** (Mistral OCR API compatible):
* ```bash
* curl -X POST "http://localhost:4000/v1/ocr" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -H "Content-Type: application/json" -d '{
* "model": "mistral-ocr",
* "document": {
* "type": "document_url",
* "document_url": "https://arxiv.org/pdf/2201.04234"
* }
* }'
* ```
*
* **2. Multipart form file upload**:
* ```bash
* curl -X POST "http://localhost:4000/v1/ocr" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -F "model=mistral-ocr" -F "[email protected]"
* ```
*
* Response format is normalized to the LiteLLM OCR schema by default. Providers
* that support it (Azure Document Intelligence) can return their own payload
* instead, with cost tracking unchanged, via `x-req-format: native` (or
* `"req_format": "native"` in the body).
*/
post: operations["ocr_ocr_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/onboarding/claim_token": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Claim Onboarding Link
* @description Special route. Allows UI link share user to update their password.
*
* - Get the invite link
* - Validate it's still 'valid'
* - Check if user within initial session (prevents abuse)
* - Get user from db
* - Update user password
*
* This route can only update user password.
*/
post: operations["claim_onboarding_link_onboarding_claim_token_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/onboarding/get_token": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Onboarding
* @description - Get the invite link
* - Validate it's still 'valid'
* - Return a short-lived onboarding token
* - Get user from db
* - Pass in user_email if set
*/
get: operations["onboarding_onboarding_get_token_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/openai/deployments/{model}/chat/completions": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Chat Completion
* @description Follows the exact same API spec as `OpenAI's Chat API https://platform.openai.com/docs/api-reference/chat`
*
* ```bash
* curl -X POST http://localhost:4000/v1/chat/completions
* -H "Content-Type: application/json"
* -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* -d '{
* "model": "gpt-4o",
* "messages": [
* {
* "role": "user",
* "content": "Hello!"
* }
* ]
* }'
* ```
*/
post: operations["chat_completion_openai_deployments__model__chat_completions_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/openai/deployments/{model}/completions": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Completion
* @description Follows the exact same API spec as `OpenAI's Completions API https://platform.openai.com/docs/api-reference/completions`
*
* ```bash
* curl -X POST http://localhost:4000/v1/completions
* -H "Content-Type: application/json"
* -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* -d '{
* "model": "gpt-3.5-turbo-instruct",
* "prompt": "Once upon a time",
* "max_tokens": 50,
* "temperature": 0.7
* }'
* ```
*/
post: operations["completion_openai_deployments__model__completions_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/openai/deployments/{model}/embeddings": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Embeddings
* @description Follows the exact same API spec as `OpenAI's Embeddings API https://platform.openai.com/docs/api-reference/embeddings`
*
* ```bash
* curl -X POST http://localhost:4000/v1/embeddings
* -H "Content-Type: application/json"
* -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* -d '{
* "model": "text-embedding-ada-002",
* "input": "The quick brown fox jumps over the lazy dog"
* }'
* ```
*/
post: operations["embeddings_openai_deployments__model__embeddings_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/openai/deployments/{model}/images/edits": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Image Edit Api
* @description Follows the OpenAI Images API spec: https://platform.openai.com/docs/api-reference/images/create
*
* ```bash
* curl -s -D >(grep -i x-request-id >&2) -o >(jq -r '.data[0].b64_json' | base64 --decode > gift-basket.png) -X POST "http://localhost:4000/v1/images/edits" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -F "model=gpt-image-1" -F "image[][email protected]" -F 'prompt=Create a studio ghibli image of this'
* ```
*/
post: operations["image_edit_api_openai_deployments__model__images_edits_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/openai/deployments/{model}/images/generations": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/** Image Generation */
post: operations["image_generation_openai_deployments__model__images_generations_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/openai/v1/live/sessions": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* WebSocket: live_sessions_websocket_endpoint
* @description WebSocket connection endpoint
*/
get: operations["websocket_live_sessions_websocket_endpoint_get_3"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/openai/v1/realtime": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* WebSocket: realtime_websocket_endpoint
* @description WebSocket connection endpoint
*/
get: operations["websocket_realtime_websocket_endpoint_get_3"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/openai/v1/realtime/calls": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/** Proxy Realtime Calls */
post: operations["proxy_realtime_calls_openai_v1_realtime_calls_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/openai/v1/realtime/client_secrets": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/** Create Realtime Client Secret */
post: operations["create_realtime_client_secret_openai_v1_realtime_client_secrets_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/openai/v1/realtime/transcription_sessions": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Create Realtime Transcription Session
* @description Create an ephemeral Realtime transcription session
* (POST /v1/realtime/transcription_sessions) for the WebRTC/WebSocket flow.
*
* Mirrors the client_secrets route but targets the transcription_sessions
* endpoint and encrypts the ephemeral key returned under `client_secret.value`.
*/
post: operations["create_realtime_transcription_session_openai_v1_realtime_transcription_sessions_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/openai/v1/responses": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Responses Api
* @description Follows the OpenAI Responses API spec: https://platform.openai.com/docs/api-reference/responses
*
* Supports background mode with polling_via_cache for partial response retrieval.
* When background=true and polling_via_cache is enabled, returns a polling_id immediately
* and streams the response in the background, updating Redis cache.
*
* ```bash
* # Normal request
* curl -X POST http://localhost:4000/v1/responses -H "Content-Type: application/json" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -d '{
* "model": "gpt-4o",
* "input": "Tell me about AI"
* }'
*
* # Background request with polling
* curl -X POST http://localhost:4000/v1/responses -H "Content-Type: application/json" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -d '{
* "model": "gpt-4o",
* "input": "Tell me about AI",
* "background": true
* }'
* ```
*/
post: operations["responses_api_openai_v1_responses_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/openai/v1/responses/compact": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Compact Response
* @description Compact a response by running a compaction pass over a conversation.
*
* Returns encrypted, opaque items that can be used to reduce context size.
*
* Follows the OpenAI Responses API spec: https://platform.openai.com/docs/api-reference/responses/compact
*
* ```bash
* curl -X POST http://localhost:4000/v1/responses/compact -H "Content-Type: application/json" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -d '{
* "model": "gpt-4o",
* "input": [{"role": "user", "content": "Hello"}]
* }'
* ```
*/
post: operations["compact_response_openai_v1_responses_compact_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/openai/v1/responses/input_tokens": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Responses Input Tokens
* @description Count the input tokens of a Responses API request without calling the model.
*
* Follows the OpenAI Responses API spec: https://platform.openai.com/docs/api-reference/responses/input-tokens
*
* ```bash
* curl -X POST http://localhost:4000/v1/responses/input_tokens -H "Content-Type: application/json" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -d '{
* "model": "gpt-4o",
* "input": "Hello, how are you?"
* }'
* ```
*
* Returns: `{"object": "response.input_tokens", "input_tokens": <count>}`
*/
post: operations["responses_input_tokens_openai_v1_responses_input_tokens_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/openai/v1/responses/{response_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Response
* @description Get a response by ID.
*
* Supports both:
* - Polling IDs (litellm_poll_*): Returns cumulative cached content from background responses
* - Provider response IDs: Passes through to provider API
*
* Follows the OpenAI Responses API spec: https://platform.openai.com/docs/api-reference/responses/get
*
* ```bash
* # Get polling response
* curl -X GET http://localhost:4000/v1/responses/litellm_poll_abc123 -H "Authorization: Bearer $LITELLM_MASTER_KEY"
*
* # Get provider response
* curl -X GET http://localhost:4000/v1/responses/resp_abc123 -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*/
get: operations["get_response_openai_v1_responses__response_id__get"];
put?: never;
post?: never;
/**
* Delete Response
* @description Delete a response by ID.
*
* Supports both:
* - Polling IDs (litellm_poll_*): Deletes from Redis cache
* - Provider response IDs: Passes through to provider API
*
* Follows the OpenAI Responses API spec: https://platform.openai.com/docs/api-reference/responses/delete
*
* ```bash
* curl -X DELETE http://localhost:4000/v1/responses/resp_abc123 -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*/
delete: operations["delete_response_openai_v1_responses__response_id__delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/openai/v1/responses/{response_id}/cancel": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Cancel Response
* @description Cancel a response by ID.
*
* Supports both:
* - Polling IDs (litellm_poll_*): Cancels background response and updates status in Redis
* - Provider response IDs: Passes through to provider API
*
* Follows the OpenAI Responses API spec: https://platform.openai.com/docs/api-reference/responses/cancel
*
* ```bash
* # Cancel polling response
* curl -X POST http://localhost:4000/v1/responses/litellm_poll_abc123/cancel -H "Authorization: Bearer $LITELLM_MASTER_KEY"
*
* # Cancel provider response
* curl -X POST http://localhost:4000/v1/responses/resp_abc123/cancel -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*/
post: operations["cancel_response_openai_v1_responses__response_id__cancel_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/openai/v1/responses/{response_id}/input_items": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Response Input Items
* @description List input items for a response.
*/
get: operations["get_response_input_items_openai_v1_responses__response_id__input_items_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/openai/{endpoint}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Openai Proxy Route
* @description Pass-through endpoint for OpenAI API calls.
*
* Available on both routes:
* - /openai/{endpoint:path} - Standard OpenAI passthrough route
* - /openai_passthrough/{endpoint:path} - Dedicated passthrough route (recommended for Responses API)
*
* Use /openai_passthrough/* when you need guaranteed passthrough to OpenAI without conflicts
* with LiteLLM's native implementations (e.g., for the Responses API at /v1/responses).
*
* Examples:
* Standard route:
* - /openai/v1/chat/completions
* - /openai/v1/assistants
* - /openai/v1/threads
*
* Dedicated passthrough (for Responses API):
* - /openai_passthrough/v1/responses
* - /openai_passthrough/v1/responses/{response_id}
* - /openai_passthrough/v1/responses/{response_id}/input_items
*
* [Docs](https://docs.litellm.ai/docs/pass_through/openai_passthrough)
*/
get: operations["openai_proxy_route_openai__endpoint__get"];
/**
* Openai Proxy Route
* @description Pass-through endpoint for OpenAI API calls.
*
* Available on both routes:
* - /openai/{endpoint:path} - Standard OpenAI passthrough route
* - /openai_passthrough/{endpoint:path} - Dedicated passthrough route (recommended for Responses API)
*
* Use /openai_passthrough/* when you need guaranteed passthrough to OpenAI without conflicts
* with LiteLLM's native implementations (e.g., for the Responses API at /v1/responses).
*
* Examples:
* Standard route:
* - /openai/v1/chat/completions
* - /openai/v1/assistants
* - /openai/v1/threads
*
* Dedicated passthrough (for Responses API):
* - /openai_passthrough/v1/responses
* - /openai_passthrough/v1/responses/{response_id}
* - /openai_passthrough/v1/responses/{response_id}/input_items
*
* [Docs](https://docs.litellm.ai/docs/pass_through/openai_passthrough)
*/
put: operations["openai_proxy_route_openai__endpoint__put"];
/**
* Openai Proxy Route
* @description Pass-through endpoint for OpenAI API calls.
*
* Available on both routes:
* - /openai/{endpoint:path} - Standard OpenAI passthrough route
* - /openai_passthrough/{endpoint:path} - Dedicated passthrough route (recommended for Responses API)
*
* Use /openai_passthrough/* when you need guaranteed passthrough to OpenAI without conflicts
* with LiteLLM's native implementations (e.g., for the Responses API at /v1/responses).
*
* Examples:
* Standard route:
* - /openai/v1/chat/completions
* - /openai/v1/assistants
* - /openai/v1/threads
*
* Dedicated passthrough (for Responses API):
* - /openai_passthrough/v1/responses
* - /openai_passthrough/v1/responses/{response_id}
* - /openai_passthrough/v1/responses/{response_id}/input_items
*
* [Docs](https://docs.litellm.ai/docs/pass_through/openai_passthrough)
*/
post: operations["openai_proxy_route_openai__endpoint__post"];
/**
* Openai Proxy Route
* @description Pass-through endpoint for OpenAI API calls.
*
* Available on both routes:
* - /openai/{endpoint:path} - Standard OpenAI passthrough route
* - /openai_passthrough/{endpoint:path} - Dedicated passthrough route (recommended for Responses API)
*
* Use /openai_passthrough/* when you need guaranteed passthrough to OpenAI without conflicts
* with LiteLLM's native implementations (e.g., for the Responses API at /v1/responses).
*
* Examples:
* Standard route:
* - /openai/v1/chat/completions
* - /openai/v1/assistants
* - /openai/v1/threads
*
* Dedicated passthrough (for Responses API):
* - /openai_passthrough/v1/responses
* - /openai_passthrough/v1/responses/{response_id}
* - /openai_passthrough/v1/responses/{response_id}/input_items
*
* [Docs](https://docs.litellm.ai/docs/pass_through/openai_passthrough)
*/
delete: operations["openai_proxy_route_openai__endpoint__delete"];
options?: never;
head?: never;
/**
* Openai Proxy Route
* @description Pass-through endpoint for OpenAI API calls.
*
* Available on both routes:
* - /openai/{endpoint:path} - Standard OpenAI passthrough route
* - /openai_passthrough/{endpoint:path} - Dedicated passthrough route (recommended for Responses API)
*
* Use /openai_passthrough/* when you need guaranteed passthrough to OpenAI without conflicts
* with LiteLLM's native implementations (e.g., for the Responses API at /v1/responses).
*
* Examples:
* Standard route:
* - /openai/v1/chat/completions
* - /openai/v1/assistants
* - /openai/v1/threads
*
* Dedicated passthrough (for Responses API):
* - /openai_passthrough/v1/responses
* - /openai_passthrough/v1/responses/{response_id}
* - /openai_passthrough/v1/responses/{response_id}/input_items
*
* [Docs](https://docs.litellm.ai/docs/pass_through/openai_passthrough)
*/
patch: operations["openai_proxy_route_openai__endpoint__patch"];
trace?: never;
};
"/openai_passthrough/{endpoint}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Openai Passthrough Route
* @description Dedicated pass-through to the OpenAI API with no overlap with LiteLLM's native
* implementations (e.g. the Responses API at /v1/responses).
*
* Examples:
* - /openai_passthrough/v1/responses
* - /openai_passthrough/v1/responses/{response_id}
* - /openai_passthrough/v1/responses/{response_id}/input_items
*
* [Docs](https://docs.litellm.ai/docs/pass_through/openai_passthrough)
*/
get: operations["openai_passthrough_route_openai_passthrough__endpoint__get"];
/**
* Openai Passthrough Route
* @description Dedicated pass-through to the OpenAI API with no overlap with LiteLLM's native
* implementations (e.g. the Responses API at /v1/responses).
*
* Examples:
* - /openai_passthrough/v1/responses
* - /openai_passthrough/v1/responses/{response_id}
* - /openai_passthrough/v1/responses/{response_id}/input_items
*
* [Docs](https://docs.litellm.ai/docs/pass_through/openai_passthrough)
*/
put: operations["openai_passthrough_route_openai_passthrough__endpoint__put"];
/**
* Openai Passthrough Route
* @description Dedicated pass-through to the OpenAI API with no overlap with LiteLLM's native
* implementations (e.g. the Responses API at /v1/responses).
*
* Examples:
* - /openai_passthrough/v1/responses
* - /openai_passthrough/v1/responses/{response_id}
* - /openai_passthrough/v1/responses/{response_id}/input_items
*
* [Docs](https://docs.litellm.ai/docs/pass_through/openai_passthrough)
*/
post: operations["openai_passthrough_route_openai_passthrough__endpoint__post"];
/**
* Openai Passthrough Route
* @description Dedicated pass-through to the OpenAI API with no overlap with LiteLLM's native
* implementations (e.g. the Responses API at /v1/responses).
*
* Examples:
* - /openai_passthrough/v1/responses
* - /openai_passthrough/v1/responses/{response_id}
* - /openai_passthrough/v1/responses/{response_id}/input_items
*
* [Docs](https://docs.litellm.ai/docs/pass_through/openai_passthrough)
*/
delete: operations["openai_passthrough_route_openai_passthrough__endpoint__delete"];
options?: never;
head?: never;
/**
* Openai Passthrough Route
* @description Dedicated pass-through to the OpenAI API with no overlap with LiteLLM's native
* implementations (e.g. the Responses API at /v1/responses).
*
* Examples:
* - /openai_passthrough/v1/responses
* - /openai_passthrough/v1/responses/{response_id}
* - /openai_passthrough/v1/responses/{response_id}/input_items
*
* [Docs](https://docs.litellm.ai/docs/pass_through/openai_passthrough)
*/
patch: operations["openai_passthrough_route_openai_passthrough__endpoint__patch"];
trace?: never;
};
"/openrouter/{endpoint}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Openrouter Proxy Route */
get: operations["openrouter_proxy_route_openrouter__endpoint__get"];
/** Openrouter Proxy Route */
put: operations["openrouter_proxy_route_openrouter__endpoint__put"];
/** Openrouter Proxy Route */
post: operations["openrouter_proxy_route_openrouter__endpoint__post"];
/** Openrouter Proxy Route */
delete: operations["openrouter_proxy_route_openrouter__endpoint__delete"];
options?: never;
head?: never;
/** Openrouter Proxy Route */
patch: operations["openrouter_proxy_route_openrouter__endpoint__patch"];
trace?: never;
};
"/organization/daily/activity": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Organization Daily Activity
* @description Get daily activity for specific organizations or all accessible organizations.
*/
get: operations["get_organization_daily_activity_organization_daily_activity_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/organization/daily/activity/aggregated": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get Organization Daily Activity Aggregated */
get: operations["get_organization_daily_activity_aggregated_organization_daily_activity_aggregated_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/organization/daily/activity/aggregated/keys": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get Organization Daily Activity Aggregated Keys */
get: operations["get_organization_daily_activity_aggregated_keys_organization_daily_activity_aggregated_keys_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/organization/daily/activity/aggregated/model_top_keys": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get Organization Daily Activity Model Top Keys */
get: operations["get_organization_daily_activity_model_top_keys_organization_daily_activity_aggregated_model_top_keys_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/organization/daily/activity/aggregated/search": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get Organization Daily Activity Aggregated Search */
get: operations["get_organization_daily_activity_aggregated_search_organization_daily_activity_aggregated_search_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/organization/daily/activity/export": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get Organization Daily Activity Export */
get: operations["get_organization_daily_activity_export_organization_daily_activity_export_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/organization/delete": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
post?: never;
/**
* Delete Organization
* @description Delete an organization
*
* # Parameters:
*
* - organization_ids: List[str] - The organization ids to delete.
*/
delete: operations["delete_organization_organization_delete_delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/organization/info": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Info Organization
* @description Get the org specific information
*/
get: operations["info_organization_organization_info_get"];
put?: never;
/**
* Deprecated Info Organization
* @description DEPRECATED: Use GET /organization/info instead
*/
post: operations["deprecated_info_organization_organization_info_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/organization/list": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List Organization
* @description Get a list of organizations with optional filtering.
*
* Parameters:
* org_id: Optional[str]
* Filter organizations by exact organization_id match
* org_alias: Optional[str]
* Filter organizations by partial organization_alias match (case-insensitive)
*
* Example:
* ```
* curl --location --request GET 'http://0.0.0.0:4000/organization/list?org_alias=my-org' --header "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*
* Example with org_id:
* ```
* curl --location --request GET 'http://0.0.0.0:4000/organization/list?org_id=123e4567-e89b-12d3-a456-426614174000' --header "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*/
get: operations["list_organization_organization_list_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/organization/member_add": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Organization Member Add
* @description [BETA]
*
* Add new members (either via user_email or user_id) to an organization
*
* If user doesn't exist, new user row will also be added to User Table
*
* Only proxy_admin or org_admin of organization, allowed to access this endpoint.
*
* # Parameters:
*
* - organization_id: str (required)
* - member: Union[List[Member], Member] (required)
* - role: Literal[LitellmUserRoles] (required)
* - user_id: Optional[str]
* - user_email: Optional[str]
*
* Note: Either user_id or user_email must be provided for each member.
*
* Example:
* ```
* curl -X POST 'http://0.0.0.0:4000/organization/member_add' -H "Authorization: Bearer $LITELLM_MASTER_KEY" -H 'Content-Type: application/json' -d '{
* "organization_id": "45e3e396-ee08-4a61-a88e-16b3ce7e0849",
* "member": {
* "role": "internal_user",
* "user_id": "[email protected]"
* },
* "max_budget_in_organization": 100.0
* }'
* ```
*
* The following is executed in this function:
*
* 1. Check if organization exists
* 2. Creates a new Internal User if the user_id or user_email is not found in LiteLLM_UserTable
* 3. Add Internal User to the `LiteLLM_OrganizationMembership` table
*/
post: operations["organization_member_add_organization_member_add_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/organization/member_delete": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
post?: never;
/**
* Organization Member Delete
* @description Delete a member from an organization
*/
delete: operations["organization_member_delete_organization_member_delete_delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/organization/member_update": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
/**
* Organization Member Update
* @description Update a member's role in an organization
*/
patch: operations["organization_member_update_organization_member_update_patch"];
trace?: never;
};
"/organization/new": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* New Organization
* @description Allow orgs to own teams
*
* Set org level budgets + model access.
*
* Only admins can create orgs.
*
* # Parameters
*
* - organization_alias: *str* - The name of the organization.
* - models: *List* - The models the organization has access to.
* - budget_id: *Optional[str]* - The id for a budget (tpm/rpm/max budget) for the organization.
* ### IF NO BUDGET ID - CREATE ONE WITH THESE PARAMS ###
* - max_budget: *Optional[float]* - Max budget for org
* - tpm_limit: *Optional[int]* - Max tpm limit for org
* - rpm_limit: *Optional[int]* - Max rpm limit for org
* - tpd_limit: *Optional[int]* - Max tokens per day stored on the org budget. Batch submissions enforce tpd_limit at the key, team and end user scopes only.
* - model_rpm_limit: *Optional[Dict[str, int]]* - The RPM (Requests Per Minute) limit per model for this organization.
* - model_tpm_limit: *Optional[Dict[str, int]]* - The TPM (Tokens Per Minute) limit per model for this organization.
* - max_parallel_requests: *Optional[int]* - [Not Implemented Yet] Max parallel requests for org
* - soft_budget: *Optional[float]* - [Not Implemented Yet] Get a slack alert when this soft budget is reached. Don't block requests.
* - model_max_budget: *Optional[dict]* - Max budget for a specific model
* - budget_duration: *Optional[str]* - Frequency of reseting org budget
* - metadata: *Optional[dict]* - Metadata for organization, store information for organization. Example metadata - {"extra_info": "some info"}
* - blocked: *bool* - Flag indicating if the org is blocked or not - will stop all calls from keys with this org_id.
* - tags: *Optional[List[str]]* - Tags for [tracking spend](https://litellm.vercel.app/docs/proxy/enterprise#tracking-spend-for-custom-tags) and/or doing [tag-based routing](https://litellm.vercel.app/docs/proxy/tag_routing).
* - organization_id: *Optional[str]* - The organization id of the team. Default is None. Create via `/organization/new`.
* - model_aliases: Optional[dict] - Model aliases for the team. [Docs](https://docs.litellm.ai/docs/proxy/team_based_routing#create-team-with-model-alias)
* - object_permission: Optional[LiteLLM_ObjectPermissionBase] - organization-specific object permission. Example - {"vector_stores": ["vector_store_1", "vector_store_2"]}. IF null or {} then no object permission.
* - allowed_models: Optional[List[str]] - List of models the organization is allowed to access. If not set, defaults to the models field.
* - temp_budget_increase: *Optional[float]* - Stored on the org budget row but only enforced for team member budgets today.
* - temp_budget_expiry: *Optional[str]* - Stored on the org budget row but only enforced for team member budgets today.
* Case 1: Create new org **without** a budget_id
*
* ```bash
* curl --location 'http://0.0.0.0:4000/organization/new'
* --header "Authorization: Bearer $LITELLM_MASTER_KEY"
* --header 'Content-Type: application/json'
* --data '{
* "organization_alias": "my-secret-org",
* "models": ["model1", "model2"],
* "max_budget": 100
* }'
*
*
* ```
*
* Case 2: Create new org **with** a budget_id
*
* ```bash
* curl --location 'http://0.0.0.0:4000/organization/new'
* --header "Authorization: Bearer $LITELLM_MASTER_KEY"
* --header 'Content-Type: application/json'
* --data '{
* "organization_alias": "my-secret-org",
* "models": ["model1", "model2"],
* "budget_id": "428eeaa8-f3ac-4e85-a8fb-7dc8d7aa8689"
* }'
* ```
*/
post: operations["new_organization_organization_new_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/organization/spend/report": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Organization Spend Report
* @description Get spend for an organization over a date range, grouped by api_key with a per-model and per-team breakdown.
*
* Covers spend logged against the organization directly and against any of its
* teams. Callable by proxy admins (any organization) and org admins (their own
* organizations). Defaults to the calling key's organization_id when
* `?organization_id=` is omitted.
*/
get: operations["get_organization_spend_report_organization_spend_report_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/organization/update": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
/**
* Update Organization
* @description Update an organization
*/
patch: operations["update_organization_organization_update_patch"];
trace?: never;
};
"/otel-spans": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get Otel Spans */
get: operations["get_otel_spans_otel_spans_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/plugin-proxy/{plugin_name}/{path}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Plugin Proxy
* @description Authenticated reverse-proxy to a registered plugin backend.
*
* Restricted to proxy_admin callers — the shared plugin_key must not be
* usable as a confused-deputy credential by regular users. Plugin UIs
* talk to the plugin service directly via the iframe; this route is for
* administrative and server-to-server access only.
*
* The caller's litellm credential is stripped and replaced with the
* plugin's own plugin_key so plugins never receive a live litellm API key.
*/
get: operations["plugin_proxy_plugin_proxy__plugin_name___path__get"];
/**
* Plugin Proxy
* @description Authenticated reverse-proxy to a registered plugin backend.
*
* Restricted to proxy_admin callers — the shared plugin_key must not be
* usable as a confused-deputy credential by regular users. Plugin UIs
* talk to the plugin service directly via the iframe; this route is for
* administrative and server-to-server access only.
*
* The caller's litellm credential is stripped and replaced with the
* plugin's own plugin_key so plugins never receive a live litellm API key.
*/
put: operations["plugin_proxy_plugin_proxy__plugin_name___path__put"];
/**
* Plugin Proxy
* @description Authenticated reverse-proxy to a registered plugin backend.
*
* Restricted to proxy_admin callers — the shared plugin_key must not be
* usable as a confused-deputy credential by regular users. Plugin UIs
* talk to the plugin service directly via the iframe; this route is for
* administrative and server-to-server access only.
*
* The caller's litellm credential is stripped and replaced with the
* plugin's own plugin_key so plugins never receive a live litellm API key.
*/
post: operations["plugin_proxy_plugin_proxy__plugin_name___path__post"];
/**
* Plugin Proxy
* @description Authenticated reverse-proxy to a registered plugin backend.
*
* Restricted to proxy_admin callers — the shared plugin_key must not be
* usable as a confused-deputy credential by regular users. Plugin UIs
* talk to the plugin service directly via the iframe; this route is for
* administrative and server-to-server access only.
*
* The caller's litellm credential is stripped and replaced with the
* plugin's own plugin_key so plugins never receive a live litellm API key.
*/
delete: operations["plugin_proxy_plugin_proxy__plugin_name___path__delete"];
/**
* Plugin Proxy
* @description Authenticated reverse-proxy to a registered plugin backend.
*
* Restricted to proxy_admin callers — the shared plugin_key must not be
* usable as a confused-deputy credential by regular users. Plugin UIs
* talk to the plugin service directly via the iframe; this route is for
* administrative and server-to-server access only.
*
* The caller's litellm credential is stripped and replaced with the
* plugin's own plugin_key so plugins never receive a live litellm API key.
*/
options: operations["plugin_proxy_plugin_proxy__plugin_name___path__options"];
/**
* Plugin Proxy
* @description Authenticated reverse-proxy to a registered plugin backend.
*
* Restricted to proxy_admin callers — the shared plugin_key must not be
* usable as a confused-deputy credential by regular users. Plugin UIs
* talk to the plugin service directly via the iframe; this route is for
* administrative and server-to-server access only.
*
* The caller's litellm credential is stripped and replaced with the
* plugin's own plugin_key so plugins never receive a live litellm API key.
*/
head: operations["plugin_proxy_plugin_proxy__plugin_name___path__head"];
/**
* Plugin Proxy
* @description Authenticated reverse-proxy to a registered plugin backend.
*
* Restricted to proxy_admin callers — the shared plugin_key must not be
* usable as a confused-deputy credential by regular users. Plugin UIs
* talk to the plugin service directly via the iframe; this route is for
* administrative and server-to-server access only.
*
* The caller's litellm credential is stripped and replaced with the
* plugin's own plugin_key so plugins never receive a live litellm API key.
*/
patch: operations["plugin_proxy_plugin_proxy__plugin_name___path__patch"];
trace?: never;
};
"/policies": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Create Policy
* @description Create a new policy.
*
* Example Request:
* ```bash
* curl -X POST "http://localhost:4000/policies" \
* -H "Authorization: Bearer <your_api_key>" \
* -H "Content-Type: application/json" \
* -d '{
* "policy_name": "global-baseline",
* "description": "Base guardrails for all requests",
* "guardrails_add": ["pii_masking", "prompt_injection"],
* "guardrails_remove": []
* }'
* ```
*
* Example Response:
* ```json
* {
* "policy_id": "123e4567-e89b-12d3-a456-426614174000",
* "policy_name": "global-baseline",
* "inherit": null,
* "description": "Base guardrails for all requests",
* "guardrails_add": ["pii_masking", "prompt_injection"],
* "guardrails_remove": [],
* "condition": null,
* "created_at": "2024-01-01T00:00:00Z",
* "updated_at": "2024-01-01T00:00:00Z"
* }
* ```
*/
post: operations["create_policy_policies_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/policies/attachments": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Create Policy Attachment
* @description Create a new policy attachment.
*
* Example Request:
* ```bash
* curl -X POST "http://localhost:4000/policies/attachments" \
* -H "Authorization: Bearer <your_api_key>" \
* -H "Content-Type: application/json" \
* -d '{
* "policy_name": "global-baseline",
* "scope": "*"
* }'
* ```
*
* Example with team-specific attachment:
* ```bash
* curl -X POST "http://localhost:4000/policies/attachments" \
* -H "Authorization: Bearer <your_api_key>" \
* -H "Content-Type: application/json" \
* -d '{
* "policy_name": "healthcare-compliance",
* "teams": ["healthcare-team", "medical-research"]
* }'
* ```
*
* Example Response:
* ```json
* {
* "attachment_id": "123e4567-e89b-12d3-a456-426614174000",
* "policy_name": "global-baseline",
* "scope": "*",
* "teams": [],
* "keys": [],
* "models": [],
* "created_at": "2024-01-01T00:00:00Z",
* "updated_at": "2024-01-01T00:00:00Z"
* }
* ```
*/
post: operations["create_policy_attachment_policies_attachments_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/policies/attachments/estimate-impact": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Estimate Attachment Impact
* @description Estimate how many keys and teams would be affected by a policy attachment.
*
* Use this before creating an attachment to preview the blast radius.
*
* Example Request:
* ```bash
* curl -X POST "http://localhost:4000/policies/attachments/estimate-impact" \
* -H "Authorization: Bearer <your_api_key>" \
* -H "Content-Type: application/json" \
* -d '{
* "policy_name": "hipaa-compliance",
* "tags": ["healthcare", "health-*"]
* }'
* ```
*/
post: operations["estimate_attachment_impact_policies_attachments_estimate_impact_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/policies/attachments/list": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List Policy Attachments
* @description List all policy attachments from the database and config.yaml.
*
* Config-defined attachments are returned with definition_location "config" and a
* synthetic attachment_id ("config-<index>").
*
* Example Request:
* ```bash
* curl -X GET "http://localhost:4000/policies/attachments/list" \
* -H "Authorization: Bearer <your_api_key>"
* ```
*
* Example Response:
* ```json
* {
* "attachments": [
* {
* "attachment_id": "123e4567-e89b-12d3-a456-426614174000",
* "policy_name": "global-baseline",
* "scope": "*",
* "teams": [],
* "keys": [],
* "models": [],
* "created_at": "2024-01-01T00:00:00Z",
* "updated_at": "2024-01-01T00:00:00Z"
* }
* ],
* "total_count": 1
* }
* ```
*/
get: operations["list_policy_attachments_policies_attachments_list_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/policies/attachments/{attachment_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Policy Attachment
* @description Get a policy attachment by ID.
*
* Example Request:
* ```bash
* curl -X GET "http://localhost:4000/policies/attachments/123e4567-e89b-12d3-a456-426614174000" \
* -H "Authorization: Bearer <your_api_key>"
* ```
*/
get: operations["get_policy_attachment_policies_attachments__attachment_id__get"];
put?: never;
post?: never;
/**
* Delete Policy Attachment
* @description Delete a policy attachment.
*
* Example Request:
* ```bash
* curl -X DELETE "http://localhost:4000/policies/attachments/123e4567-e89b-12d3-a456-426614174000" \
* -H "Authorization: Bearer <your_api_key>"
* ```
*
* Example Response:
* ```json
* {
* "message": "Attachment 123e4567-e89b-12d3-a456-426614174000 deleted successfully"
* }
* ```
*/
delete: operations["delete_policy_attachment_policies_attachments__attachment_id__delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/policies/compare": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Compare Policy Versions
* @description Compare two policy versions. Query params: version_a, version_b (policy version IDs).
*/
get: operations["compare_policy_versions_policies_compare_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/policies/list": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List Policies
* @description List all policies from the database and config.yaml. Optionally filter by version_status.
*
* Config-defined policies are returned with definition_location "config" and are treated
* as production versions. On a name conflict with a production DB policy, only the DB policy
* is returned, mirroring runtime resolution where only production DB versions override config.
* A draft or published DB version does not hide the config policy, since the config version
* is still the one being enforced.
*
* Query params:
* - version_status: Optional. One of "draft", "published", "production".
* If omitted, all versions are returned.
*
* Example Request:
* ```bash
* curl -X GET "http://localhost:4000/policies/list" \
* -H "Authorization: Bearer <your_api_key>"
* curl -X GET "http://localhost:4000/policies/list?version_status=production" \
* -H "Authorization: Bearer <your_api_key>"
* ```
*
* Example Response:
* ```json
* {
* "policies": [
* {
* "policy_id": "123e4567-e89b-12d3-a456-426614174000",
* "policy_name": "global-baseline",
* "version_number": 1,
* "version_status": "production",
* "inherit": null,
* "description": "Base guardrails for all requests",
* "guardrails_add": ["pii_masking"],
* "guardrails_remove": [],
* "condition": null,
* "created_at": "2024-01-01T00:00:00Z",
* "updated_at": "2024-01-01T00:00:00Z"
* }
* ],
* "total_count": 1
* }
* ```
*/
get: operations["list_policies_policies_list_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/policies/name/{policy_name}/all-versions": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
post?: never;
/**
* Delete All Policy Versions
* @description Delete all versions of a policy. Also removes from in-memory registry.
*/
delete: operations["delete_all_policy_versions_policies_name__policy_name__all_versions_delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/policies/name/{policy_name}/versions": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List Policy Versions
* @description List all versions of a policy by name, ordered by version_number descending.
*/
get: operations["list_policy_versions_policies_name__policy_name__versions_get"];
put?: never;
/**
* Create Policy Version
* @description Create a new draft version of a policy. Copies all fields from the source.
* Source is current production if source_policy_id is not provided.
*/
post: operations["create_policy_version_policies_name__policy_name__versions_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/policies/resolve": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Resolve Policies For Context
* @description Resolve which policies and guardrails apply for a given context.
*
* Use this endpoint to debug "what guardrails would apply to a request
* with this team/key/model/tags combination?"
*
* Example Request:
* ```bash
* curl -X POST "http://localhost:4000/policies/resolve" \
* -H "Authorization: Bearer <your_api_key>" \
* -H "Content-Type: application/json" \
* -d '{
* "tags": ["healthcare"],
* "model": "gpt-4"
* }'
* ```
*/
post: operations["resolve_policies_for_context_policies_resolve_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/policies/test-pipeline": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Test Pipeline
* @description Test a guardrail pipeline with sample messages.
*
* Executes the pipeline steps against the provided test messages and returns
* step-by-step results showing which guardrails passed/failed, actions taken,
* and timing information.
*
* Example Request:
* ```bash
* curl -X POST "http://localhost:4000/policies/test-pipeline" \
* -H "Authorization: Bearer <your_api_key>" \
* -H "Content-Type: application/json" \
* -d '{
* "pipeline": {
* "mode": "pre_call",
* "steps": [
* {"guardrail": "pii-guard", "on_pass": "next", "on_fail": "block"}
* ]
* },
* "test_messages": [{"role": "user", "content": "My SSN is 123-45-6789"}]
* }'
* ```
*/
post: operations["test_pipeline_policies_test_pipeline_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/policies/usage/overview": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Policies Usage Overview
* @description Return policy performance overview for the dashboard.
*/
get: operations["policies_usage_overview_policies_usage_overview_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/policies/{policy_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Policy
* @description Get a policy by ID.
*
* Example Request:
* ```bash
* curl -X GET "http://localhost:4000/policies/123e4567-e89b-12d3-a456-426614174000" \
* -H "Authorization: Bearer <your_api_key>"
* ```
*/
get: operations["get_policy_policies__policy_id__get"];
/**
* Update Policy
* @description Update an existing policy.
*
* Example Request:
* ```bash
* curl -X PUT "http://localhost:4000/policies/123e4567-e89b-12d3-a456-426614174000" \
* -H "Authorization: Bearer <your_api_key>" \
* -H "Content-Type: application/json" \
* -d '{
* "description": "Updated description",
* "guardrails_add": ["pii_masking", "toxicity_filter"]
* }'
* ```
*/
put: operations["update_policy_policies__policy_id__put"];
post?: never;
/**
* Delete Policy
* @description Delete a policy.
*
* Example Request:
* ```bash
* curl -X DELETE "http://localhost:4000/policies/123e4567-e89b-12d3-a456-426614174000" \
* -H "Authorization: Bearer <your_api_key>"
* ```
*
* Example Response:
* ```json
* {
* "message": "Policy 123e4567-e89b-12d3-a456-426614174000 deleted successfully"
* }
* ```
*/
delete: operations["delete_policy_policies__policy_id__delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/policies/{policy_id}/resolved-guardrails": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Resolved Guardrails
* @description Get the resolved guardrails for a policy (including inherited guardrails).
*
* This endpoint resolves the full inheritance chain and returns the final
* set of guardrails that would be applied for this policy.
*
* Example Request:
* ```bash
* curl -X GET "http://localhost:4000/policies/123e4567-e89b-12d3-a456-426614174000/resolved-guardrails" \
* -H "Authorization: Bearer <your_api_key>"
* ```
*
* Example Response:
* ```json
* {
* "policy_id": "123e4567-e89b-12d3-a456-426614174000",
* "policy_name": "healthcare-compliance",
* "resolved_guardrails": ["pii_masking", "prompt_injection", "toxicity_filter"]
* }
* ```
*/
get: operations["get_resolved_guardrails_policies__policy_id__resolved_guardrails_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/policies/{policy_id}/status": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
/**
* Update Policy Version Status
* @description Update a policy version's status. Valid transitions:
* - draft -> published
* - published -> production (demotes current production to published)
* - production -> published (demotes, policy becomes inactive)
*/
put: operations["update_policy_version_status_policies__policy_id__status_put"];
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/policy/info/{policy_name}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Policy Info
* @description Get detailed information about a specific policy.
*
* Returns:
* - Policy configuration
* - Resolved guardrails (after inheritance)
* - Inheritance chain
*/
get: operations["get_policy_info_policy_info__policy_name__get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/policy/list": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List Policies
* @description List all loaded policies with their resolved guardrails.
*
* Returns information about each policy including:
* - Inheritance configuration
* - Scope (teams, keys, models)
* - Guardrails to add/remove
* - Resolved guardrails (after inheritance)
* - Inheritance chain
*/
get: operations["list_policies_policy_list_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/policy/templates": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Policy Templates
* @description Get policy templates for the UI (pre-configured guardrail combinations).
*
* Fetches from GitHub with automatic fallback to local backup on failure.
* Set LITELLM_LOCAL_POLICY_TEMPLATES=true to skip GitHub and use local backup only.
*/
get: operations["get_policy_templates_policy_templates_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/policy/templates/enrich": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Enrich Policy Template
* @description Enrich a policy template with LLM-discovered data (e.g. competitor names).
*
* Calls an onboarded LLM to discover competitors for the given brand name,
* then returns enriched guardrailDefinitions with the discovered data populated.
*/
post: operations["enrich_policy_template_policy_templates_enrich_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/policy/templates/enrich/stream": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Enrich Policy Template Stream
* @description Stream competitor names as SSE events as the LLM generates them.
*
* Events:
* - data: {"type": "competitor", "name": "..."} — each competitor as discovered
* - data: {"type": "done", "competitors": [...], "competitor_variations": {...}, "guardrailDefinitions": [...]}
*/
post: operations["enrich_policy_template_stream_policy_templates_enrich_stream_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/policy/templates/suggest": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Suggest Policy Templates
* @description Use AI to suggest policy templates based on attack examples and descriptions.
*
* Calls an LLM with tool calling to match user requirements to available templates.
*/
post: operations["suggest_policy_templates_policy_templates_suggest_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/policy/templates/test": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Test Policy Template
* @description Test a policy template's guardrails against a text input without creating them.
*
* Instantiates temporary guardrails from the template definitions, runs them
* against the provided text, and returns per-guardrail results so users can
* verify the template solves their problem before creating it.
*/
post: operations["test_policy_template_policy_templates_test_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/policy/test": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Test Policy Matching
* @description Test which policies would match a given request context.
*
* This is useful for debugging and understanding policy behavior.
*
* Request body:
* ```json
* {
* "team_alias": "healthcare-team",
* "key_alias": "my-api-key",
* "model": "gpt-4"
* }
* ```
*
* Returns:
* - matching_policies: List of policy names that match
* - resolved_guardrails: Final list of guardrails that would be applied
*/
post: operations["test_policy_matching_policy_test_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/policy/validate": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Validate Policy
* @description Validate a policy configuration before applying it.
*
* Checks:
* - All referenced guardrails exist in the guardrail registry
* - All non-wildcard team aliases exist in the database
* - All non-wildcard key aliases exist in the database
* - Inheritance chains are valid (no cycles, parents exist)
* - Scope patterns are syntactically valid
*
* Returns:
* - valid: True if the policy configuration is valid (no blocking errors)
* - errors: List of blocking validation errors
* - warnings: List of non-blocking validation warnings
*
* Example request:
* ```json
* {
* "policies": {
* "global-baseline": {
* "guardrails": {
* "add": ["pii_blocker", "phi_blocker"]
* },
* "scope": {
* "teams": ["*"],
* "keys": ["*"],
* "models": ["*"]
* }
* },
* "healthcare-compliance": {
* "inherit": "global-baseline",
* "guardrails": {
* "add": ["hipaa_audit"]
* },
* "scope": {
* "teams": ["healthcare-team"]
* }
* }
* }
* }
* ```
*/
post: operations["validate_policy_policy_validate_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/project/delete": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
post?: never;
/**
* Delete Project
* @description Delete projects
*
* Parameters:
* - project_ids: *List[str]* - List of project ids to delete
*
* Example:
* ```bash
* curl --location --request DELETE 'http://0.0.0.0:4000/project/delete' \
* --header "Authorization: Bearer $LITELLM_MASTER_KEY" \
* --header 'Content-Type: application/json' \
* --data '{
* "project_ids": ["project-123", "project-456"]
* }'
* ```
*/
delete: operations["delete_project_project_delete_delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/project/info": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Project Info
* @description Get information about a specific project
*
* Parameters:
* - project_id: *str* - The project id to fetch info for
*
* Example:
* ```bash
* curl --location 'http://0.0.0.0:4000/project/info?project_id=project-123' \
* --header "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*/
get: operations["project_info_project_info_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/project/list": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List Projects
* @description List all projects that the user has access to
*
* Example:
* ```bash
* curl --location 'http://0.0.0.0:4000/project/list' \
* --header "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*/
get: operations["list_projects_project_list_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/project/new": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* New Project
* @description Create a new project. Projects sit between teams and keys in the hierarchy.
*
* Only admins or team admins can create projects.
*
* # Parameters
*
* - project_alias: *Optional[str]* - The name of the project.
* - description: *Optional[str]* - Description of the project's purpose and use case.
* - team_id: *str* - The team id that this project belongs to. Required.
* - models: *List* - The models the project has access to.
* - budget_id: *Optional[str]* - The id for a budget (tpm/rpm/max budget) for the project.
* ### IF NO BUDGET ID - CREATE ONE WITH THESE PARAMS ###
* - max_budget: *Optional[float]* - Max budget for project
* - tpm_limit: *Optional[int]* - Max tpm limit for project
* - rpm_limit: *Optional[int]* - Max rpm limit for project
* - max_parallel_requests: *Optional[int]* - Max parallel requests for project
* - soft_budget: *Optional[float]* - Get a slack alert when this soft budget is reached. Don't block requests.
* - model_max_budget: *Optional[dict]* - Max budget for a specific model. Example: {"gpt-4": 100.0, "gpt-3.5-turbo": 50.0}
* - model_rpm_limit: *Optional[dict]* - RPM limits per model. Example: {"gpt-4": 1000, "gpt-3.5-turbo": 5000}
* - model_tpm_limit: *Optional[dict]* - TPM limits per model. Example: {"gpt-4": 50000, "gpt-3.5-turbo": 100000}
* - budget_duration: *Optional[str]* - Frequency of reseting project budget
* - metadata: *Optional[dict]* - Metadata for project, store information for project. Example metadata - {"use_case_id": "SNOW-12345", "responsible_ai_id": "RAI-67890"}
* - tags: *Optional[list]* - Tags for the project. Example: ["production", "api"]
* - blocked: *bool* - Flag indicating if the project is blocked or not - will stop all calls from keys with this project_id.
* - object_permission: Optional[LiteLLM_ObjectPermissionBase] - project-specific object permission. Example - {"vector_stores": ["vector_store_1", "vector_store_2"]}. IF null or {} then no object permission.
*
* Example 1: Create new project **without** a budget_id, with model-specific limits
*
* ```bash
* curl --location 'http://0.0.0.0:4000/project/new' \
* --header "Authorization: Bearer $LITELLM_MASTER_KEY" \
* --header 'Content-Type: application/json' \
* --data '{
* "project_alias": "flight-search-assistant",
* "description": "AI-powered flight search and booking assistant",
* "team_id": "team-123",
* "models": ["gpt-4", "gpt-3.5-turbo"],
* "max_budget": 100,
* "model_rpm_limit": {
* "gpt-4": 1000,
* "gpt-3.5-turbo": 5000
* },
* "model_tpm_limit": {
* "gpt-4": 50000,
* "gpt-3.5-turbo": 100000
* },
* "metadata": {
* "use_case_id": "SNOW-12345",
* "responsible_ai_id": "RAI-67890"
* }
* }'
* ```
*
* Example 2: Create new project **with** a budget_id
*
* ```bash
* curl --location 'http://0.0.0.0:4000/project/new' \
* --header "Authorization: Bearer $LITELLM_MASTER_KEY" \
* --header 'Content-Type: application/json' \
* --data '{
* "project_alias": "hotel-recommendations",
* "description": "Personalized hotel recommendation engine",
* "team_id": "team-123",
* "models": ["claude-3-sonnet"],
* "budget_id": "428eeaa8-f3ac-4e85-a8fb-7dc8d7aa8689",
* "metadata": {
* "use_case_id": "SNOW-54321"
* }
* }'
* ```
*/
post: operations["new_project_project_new_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/project/update": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Update Project
* @description Update a project
*
* Parameters:
* - project_id: *str* - The project id to update. Required.
* - project_alias: *Optional[str]* - Updated name for the project
* - description: *Optional[str]* - Updated description for the project
* - team_id: *Optional[str]* - Updated team_id for the project
* - metadata: *Optional[dict]* - Updated metadata for project
* - models: *Optional[list]* - Updated list of models for the project
* - blocked: *Optional[bool]* - Updated blocked status
* - max_budget: *Optional[float]* - Updated max budget
* - tpm_limit: *Optional[int]* - Updated tpm limit
* - rpm_limit: *Optional[int]* - Updated rpm limit
* - model_rpm_limit: *Optional[dict]* - Updated RPM limits per model
* - model_tpm_limit: *Optional[dict]* - Updated TPM limits per model
* - budget_duration: *Optional[str]* - Updated budget duration
* - tags: *Optional[list]* - Updated list of tags for the project
* - object_permission: Optional[LiteLLM_ObjectPermissionBase] - Updated object permission
*
* Example:
* ```bash
* curl --location 'http://0.0.0.0:4000/project/update' \
* --header "Authorization: Bearer $LITELLM_MASTER_KEY" \
* --header 'Content-Type: application/json' \
* --data '{
* "project_id": "project-123",
* "description": "Updated flight search system with enhanced capabilities",
* "max_budget": 200,
* "model_rpm_limit": {
* "gpt-4": 2000,
* "gpt-3.5-turbo": 10000
* },
* "metadata": {
* "use_case_id": "SNOW-12345",
* "status": "active"
* }
* }'
* ```
*/
post: operations["update_project_project_update_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/prompts": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Create Prompt
* @description Create a new prompt
*
* 👉 [Prompt docs](https://docs.litellm.ai/docs/proxy/prompt_management)
*
* Example Request:
* ```bash
* curl -X POST "http://localhost:4000/prompts" \
* -H "Authorization: Bearer <your_api_key>" \
* -H "Content-Type: application/json" \
* -d '{
* "prompt_id": "my_prompt",
* "litellm_params": {
* "prompt_id": "my_prompt",
* "prompt_integration": "dotprompt",
* "prompt_data": {"content": "This is a prompt", "metadata": {"model": "gpt-4"}}
* },
* "prompt_info": {
* "prompt_type": "config"
* }
* }'
* ```
*/
post: operations["create_prompt_prompts_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/prompts/list": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List Prompts
* @description List the prompts that are available on the proxy server
*
* 👉 [Prompt docs](https://docs.litellm.ai/docs/proxy/prompt_management)
*
* Example Request:
* ```bash
* curl -X GET "http://localhost:4000/prompts/list" -H "Authorization: Bearer <your_api_key>"
* ```
*
* Example Response:
* ```json
* {
* "prompts": [
* {
* "prompt_id": "my_prompt_id",
* "litellm_params": {
* "prompt_id": "my_prompt_id",
* "prompt_integration": "dotprompt",
* "prompt_directory": "/path/to/prompts"
* },
* "prompt_info": {
* "prompt_type": "config"
* },
* "created_at": "2023-11-09T12:34:56.789Z",
* "updated_at": "2023-11-09T12:34:56.789Z"
* }
* ]
* }
* ```
*/
get: operations["list_prompts_prompts_list_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/prompts/test": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Test Prompt
* @description Test a prompt by rendering it with variables and executing an LLM call.
*
* This endpoint allows testing prompts before saving them to the database.
* The response is always streamed.
*
* 👉 [Prompt docs](https://docs.litellm.ai/docs/proxy/prompt_management)
*
* Example Request:
* ```bash
* curl -X POST "http://localhost:4000/prompts/test" \
* -H "Authorization: Bearer <your_api_key>" \
* -H "Content-Type: application/json" \
* -d '{
* "dotprompt_content": "---\nmodel: gpt-4o\ntemperature: 0.7\n---\n\nUser: Hello {{name}}",
* "prompt_variables": {
* "name": "World"
* }
* }'
* ```
*/
post: operations["test_prompt_prompts_test_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/prompts/{prompt_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Prompt Info
* @description Get detailed information about a specific prompt by ID, including prompt content
*
* 👉 [Prompt docs](https://docs.litellm.ai/docs/proxy/prompt_management)
*
* Example Request:
* ```bash
* curl -X GET "http://localhost:4000/prompts/my_prompt_id/info" \
* -H "Authorization: Bearer <your_api_key>"
* ```
*
* Example Response:
* ```json
* {
* "prompt_id": "my_prompt_id",
* "litellm_params": {
* "prompt_id": "my_prompt_id",
* "prompt_integration": "dotprompt",
* "prompt_directory": "/path/to/prompts"
* },
* "prompt_info": {
* "prompt_type": "config"
* },
* "created_at": "2023-11-09T12:34:56.789Z",
* "updated_at": "2023-11-09T12:34:56.789Z",
* "content": "System: You are a helpful assistant.
*
* User: {{user_message}}"
* }
* ```
*/
get: operations["get_prompt_info_prompts__prompt_id__get"];
/**
* Update Prompt
* @description Update an existing prompt
*
* 👉 [Prompt docs](https://docs.litellm.ai/docs/proxy/prompt_management)
*
* Example Request:
* ```bash
* curl -X PUT "http://localhost:4000/prompts/my_prompt_id" \
* -H "Authorization: Bearer <your_api_key>" \
* -H "Content-Type: application/json" \
* -d '{
* "prompt_id": "my_prompt",
* "litellm_params": {
* "prompt_id": "my_prompt",
* "prompt_integration": "dotprompt",
* "prompt_directory": "/path/to/prompts"
* },
* "prompt_info": {
* "prompt_type": "config"
* }
* }
* }'
* ```
*/
put: operations["update_prompt_prompts__prompt_id__put"];
post?: never;
/**
* Delete Prompt
* @description Delete a prompt
*
* 👉 [Prompt docs](https://docs.litellm.ai/docs/proxy/prompt_management)
*
* Example Request:
* ```bash
* curl -X DELETE "http://localhost:4000/prompts/my_prompt_id" \
* -H "Authorization: Bearer <your_api_key>"
* ```
*
* Example Response:
* ```json
* {
* "message": "Prompt my_prompt_id deleted successfully"
* }
* ```
*/
delete: operations["delete_prompt_prompts__prompt_id__delete"];
options?: never;
head?: never;
/**
* Patch Prompt
* @description Partially update an existing prompt
*
* 👉 [Prompt docs](https://docs.litellm.ai/docs/proxy/prompt_management)
*
* This endpoint allows updating specific fields of a prompt without sending the entire object.
* Only the following fields can be updated:
* - litellm_params: LiteLLM parameters for the prompt
* - prompt_info: Additional information about the prompt
*
* Example Request:
* ```bash
* curl -X PATCH "http://localhost:4000/prompts/my_prompt_id" \
* -H "Authorization: Bearer <your_api_key>" \
* -H "Content-Type: application/json" \
* -d '{
* "prompt_info": {
* "prompt_type": "db"
* }
* }'
* ```
*/
patch: operations["patch_prompt_prompts__prompt_id__patch"];
trace?: never;
};
"/prompts/{prompt_id}/info": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Prompt Info
* @description Get detailed information about a specific prompt by ID, including prompt content
*
* 👉 [Prompt docs](https://docs.litellm.ai/docs/proxy/prompt_management)
*
* Example Request:
* ```bash
* curl -X GET "http://localhost:4000/prompts/my_prompt_id/info" \
* -H "Authorization: Bearer <your_api_key>"
* ```
*
* Example Response:
* ```json
* {
* "prompt_id": "my_prompt_id",
* "litellm_params": {
* "prompt_id": "my_prompt_id",
* "prompt_integration": "dotprompt",
* "prompt_directory": "/path/to/prompts"
* },
* "prompt_info": {
* "prompt_type": "config"
* },
* "created_at": "2023-11-09T12:34:56.789Z",
* "updated_at": "2023-11-09T12:34:56.789Z",
* "content": "System: You are a helpful assistant.
*
* User: {{user_message}}"
* }
* ```
*/
get: operations["get_prompt_info_prompts__prompt_id__info_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/prompts/{prompt_id}/versions": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Prompt Versions
* @description Get all versions of a specific prompt by base prompt ID
*
* 👉 [Prompt docs](https://docs.litellm.ai/docs/proxy/prompt_management)
*
* Example Request:
* ```bash
* curl -X GET "http://localhost:4000/prompts/jack_success/versions" \
* -H "Authorization: Bearer <your_api_key>"
* ```
*
* Example Response:
* ```json
* {
* "prompts": [
* {
* "prompt_id": "jack_success.v1",
* "litellm_params": {...},
* "prompt_info": {"prompt_type": "db"},
* "created_at": "2023-11-09T12:34:56.789Z",
* "updated_at": "2023-11-09T12:34:56.789Z"
* },
* {
* "prompt_id": "jack_success.v2",
* "litellm_params": {...},
* "prompt_info": {"prompt_type": "db"},
* "created_at": "2023-11-09T13:45:12.345Z",
* "updated_at": "2023-11-09T13:45:12.345Z"
* }
* ]
* }
* ```
*/
get: operations["get_prompt_versions_prompts__prompt_id__versions_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/provider/budgets": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Provider Budgets
* @description Provider Budget Routing - Get Budget, Spend Details https://docs.litellm.ai/docs/proxy/provider_budget_routing
*
* Use this endpoint to check current budget, spend and budget reset time for a provider
*
* Example Request
*
* ```bash
* curl -X GET http://localhost:4000/provider/budgets -H "Content-Type: application/json" -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*
* Example Response
*
* ```json
* {
* "providers": {
* "openai": {
* "budget_limit": 1e-12,
* "time_period": "1d",
* "spend": 0.0,
* "budget_reset_at": null
* },
* "azure": {
* "budget_limit": 100.0,
* "time_period": "1d",
* "spend": 0.0,
* "budget_reset_at": null
* },
* "anthropic": {
* "budget_limit": 100.0,
* "time_period": "10d",
* "spend": 0.0,
* "budget_reset_at": null
* },
* "vertex_ai": {
* "budget_limit": 100.0,
* "time_period": "12d",
* "spend": 0.0,
* "budget_reset_at": null
* }
* }
* }
* ```
*/
get: operations["provider_budgets_provider_budgets_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/public/agent_hub": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get Agents */
get: operations["get_agents_public_agent_hub_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/public/agents/fields": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Agent Fields
* @description Return agent type metadata required by the dashboard create-agent flow.
*
* If an agent has `inherit_credentials_from_provider`, the provider's credential
* fields are automatically appended to the agent's credential_fields.
*
* Reads from the bundled local files. Result is cached in-process for the
* lifetime of the server process.
*/
get: operations["get_agent_fields_public_agents_fields_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/public/autorouter_presets": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Public Autorouter Presets
* @description Return the auto-router preset catalog the dashboard's template picker renders.
*
* Resolved once per process, like the model cost map: fetched from ``litellm.autorouter_presets_url``
* (override with ``LITELLM_AUTOROUTER_PRESETS_URL``) on the first request, falling back to the
* catalog bundled with the package on any failure. Set ``LITELLM_LOCAL_AUTOROUTER_PRESETS=True``
* to serve the bundled catalog only. A restart picks up a newly published catalog.
*/
get: operations["get_public_autorouter_presets_public_autorouter_presets_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/public/complexity_router/fuse_presets": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get Public Fuse Presets */
get: operations["get_public_fuse_presets_public_complexity_router_fuse_presets_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/public/complexity_router/scorer_defaults": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Complexity Scorer Defaults
* @description Return the complexity router's shipped heuristic scorer defaults, for the dashboard to prefill with.
*/
get: operations["get_complexity_scorer_defaults_public_complexity_router_scorer_defaults_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/public/endpoints": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Supported Endpoints
* @description Return the list of LiteLLM proxy endpoints and which providers support each one.
*
* Reads from the bundled local backup file. Result is cached in-process for
* the lifetime of the server process.
*/
get: operations["get_supported_endpoints_public_endpoints_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/public/litellm_blog_posts": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Litellm Blog Posts
* @description Public endpoint to get the latest LiteLLM blog posts, each with its RSS categories as ``tags``.
*
* Fetches the docs RSS feed with a 1-hour in-process cache.
* Falls back to the bundled local backup on any failure.
*/
get: operations["get_litellm_blog_posts_public_litellm_blog_posts_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/public/litellm_model_cost_map": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Litellm Model Cost Map
* @description Public endpoint to get the LiteLLM model cost map.
* Returns pricing information for all supported models.
* With catalog_only=true, returns the catalog as loaded, without entries registered at runtime for proxy deployments.
*/
get: operations["get_litellm_model_cost_map_public_litellm_model_cost_map_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/public/mcp_hub": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get Mcp Servers */
get: operations["get_mcp_servers_public_mcp_hub_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/public/model_hub": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Public Model Hub */
get: operations["public_model_hub_public_model_hub_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/public/model_hub/info": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Public Model Hub Info */
get: operations["public_model_hub_info_public_model_hub_info_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/public/providers": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Supported Providers
* @description Return a sorted list of all providers supported by LiteLLM.
*/
get: operations["get_supported_providers_public_providers_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/public/providers/fields": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Provider Fields
* @description Return provider metadata required by the dashboard create-model flow.
*
* Reads from the bundled local file. Result is cached in-process for the
* lifetime of the server process.
*/
get: operations["get_provider_fields_public_providers_fields_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/public/skill_hub": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Public Skill Hub
* @description Return enabled (public) Claude Code skills — no auth required.
*/
get: operations["public_skill_hub_public_skill_hub_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/public/v1/model_hub": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Public Model Hub List
* @description The public model groups this proxy publishes, paged, sortable, searchable and
* filterable, for the public Model Hub page. No authentication.
*
* A rejected request answers with the parameters, sort fields and filter operators
* it would have accepted, so the accepted set stays discoverable from the endpoint
* itself rather than from a copy of the spec kept here.
*
* Example curl:
* ```
* curl --location --globoff 'http://0.0.0.0:4000/public/v1/model_hub?sort=-input_cost_per_token&filter[mode][in]=chat&page_size=25'
* ```
*/
get: operations["public_model_hub_list_public_v1_model_hub_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/public/v1/model_hub/{facet}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Public Model Hub Facet
* @description The distinct providers, modes or features across the published model groups, for the
* Model Hub's filter dropdowns. No authentication.
*
* Carries the same filters and search as the list route, so a dropdown offers exactly
* the values the table can show: asking for providers under `filter[mode][in]=chat`
* lists only the providers that serve a chat model.
*
* Example curl:
* ```
* curl --location --globoff 'http://0.0.0.0:4000/public/v1/model_hub/providers?filter[mode][in]=chat&page_size=50'
* ```
*/
get: operations["public_model_hub_facet_public_v1_model_hub__facet__get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/queue/chat/completions": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/** Async Queue Request */
post: operations["async_queue_request_queue_chat_completions_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/rag/ingest": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Rag Ingest
* @description RAG Ingest endpoint - all-in-one document ingestion pipeline.
*
* Supports form upload (for files) or JSON body (for URLs).
*
* ## Form upload (for files):
* ```bash
* curl -X POST "http://localhost:4000/v1/rag/ingest" \
* -H "Authorization: Bearer $LITELLM_MASTER_KEY" \
* -F file="@document.pdf" \
* -F 'ingest_options={"vector_store": {"custom_llm_provider": "openai"}}'
* ```
*
* ## JSON body (for URLs):
* ```bash
* curl -X POST "http://localhost:4000/v1/rag/ingest" \
* -H "Authorization: Bearer $LITELLM_MASTER_KEY" \
* -H "Content-Type: application/json" \
* -d '{
* "file_url": "https://example.com/document.pdf",
* "ingest_options": {"vector_store": {"custom_llm_provider": "openai"}}
* }'
* ```
*
* ## Bedrock:
* ```bash
* curl -X POST "http://localhost:4000/v1/rag/ingest" \
* -H "Authorization: Bearer $LITELLM_MASTER_KEY" \
* -F file="@document.pdf" \
* -F 'ingest_options={"vector_store": {"custom_llm_provider": "bedrock"}}'
* ```
*/
post: operations["rag_ingest_rag_ingest_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/rag/query": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Rag Query
* @description RAG Query endpoint - search vector store, optionally rerank, and generate LLM response.
*
* This endpoint:
* 1. Extracts the query from the last user message
* 2. Searches the vector store for relevant context
* 3. Optionally reranks the results
* 4. Generates an LLM response with the retrieved context
*
* ## Example Request:
* ```bash
* curl -X POST "http://localhost:4000/v1/rag/query" \
* -H "Authorization: Bearer $LITELLM_MASTER_KEY" \
* -H "Content-Type: application/json" \
* -d '{
* "model": "gpt-4o-mini",
* "messages": [{"role": "user", "content": "What is LiteLLM?"}],
* "retrieval_config": {
* "vector_store_id": "vs_abc123",
* "custom_llm_provider": "openai",
* "top_k": 5
* }
* }'
* ```
*
* ## With Reranking:
* ```bash
* curl -X POST "http://localhost:4000/v1/rag/query" \
* -H "Authorization: Bearer $LITELLM_MASTER_KEY" \
* -H "Content-Type: application/json" \
* -d '{
* "model": "gpt-4o-mini",
* "messages": [{"role": "user", "content": "What is LiteLLM?"}],
* "retrieval_config": {
* "vector_store_id": "vs_abc123",
* "custom_llm_provider": "openai",
* "top_k": 10
* },
* "rerank": {
* "enabled": true,
* "model": "cohere/rerank-english-v3.0",
* "top_n": 3
* }
* }'
* ```
*/
post: operations["rag_query_rag_query_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/realtime": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* WebSocket: realtime_websocket_endpoint
* @description WebSocket connection endpoint
*/
get: operations["websocket_realtime_websocket_endpoint_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/realtime/calls": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/** Proxy Realtime Calls */
post: operations["proxy_realtime_calls_realtime_calls_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/realtime/client_secrets": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/** Create Realtime Client Secret */
post: operations["create_realtime_client_secret_realtime_client_secrets_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/realtime/transcription_sessions": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Create Realtime Transcription Session
* @description Create an ephemeral Realtime transcription session
* (POST /v1/realtime/transcription_sessions) for the WebRTC/WebSocket flow.
*
* Mirrors the client_secrets route but targets the transcription_sessions
* endpoint and encrypts the ephemeral key returned under `client_secret.value`.
*/
post: operations["create_realtime_transcription_session_realtime_transcription_sessions_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/register": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/** Register Client */
post: operations["register_client_register_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/reload/anthropic_beta_headers": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Reload Anthropic Beta Headers
* @description ADMIN ONLY / MASTER KEY Only Endpoint
*
* Manually reload the Anthropic beta headers configuration from the remote source.
* This will fetch fresh configuration from the anthropic_beta_headers_config.json file.
*/
post: operations["reload_anthropic_beta_headers_reload_anthropic_beta_headers_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/reload/model_cost_map": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Reload Model Cost Map
* @description ADMIN ONLY / MASTER KEY Only Endpoint
*
* Manually reload the model cost map from the remote source.
* This will fetch fresh pricing data from the model_prices_and_context_window.json file.
*/
post: operations["reload_model_cost_map_reload_model_cost_map_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/rerank": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/** Rerank */
post: operations["rerank_rerank_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/responses": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* WebSocket: responses_websocket_endpoint
* @description WebSocket connection endpoint
*/
get: operations["websocket_responses_websocket_endpoint_get"];
put?: never;
/**
* Responses Api
* @description Follows the OpenAI Responses API spec: https://platform.openai.com/docs/api-reference/responses
*
* Supports background mode with polling_via_cache for partial response retrieval.
* When background=true and polling_via_cache is enabled, returns a polling_id immediately
* and streams the response in the background, updating Redis cache.
*
* ```bash
* # Normal request
* curl -X POST http://localhost:4000/v1/responses -H "Content-Type: application/json" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -d '{
* "model": "gpt-4o",
* "input": "Tell me about AI"
* }'
*
* # Background request with polling
* curl -X POST http://localhost:4000/v1/responses -H "Content-Type: application/json" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -d '{
* "model": "gpt-4o",
* "input": "Tell me about AI",
* "background": true
* }'
* ```
*/
post: operations["responses_api_responses_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/responses/compact": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Compact Response
* @description Compact a response by running a compaction pass over a conversation.
*
* Returns encrypted, opaque items that can be used to reduce context size.
*
* Follows the OpenAI Responses API spec: https://platform.openai.com/docs/api-reference/responses/compact
*
* ```bash
* curl -X POST http://localhost:4000/v1/responses/compact -H "Content-Type: application/json" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -d '{
* "model": "gpt-4o",
* "input": [{"role": "user", "content": "Hello"}]
* }'
* ```
*/
post: operations["compact_response_responses_compact_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/responses/input_tokens": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Responses Input Tokens
* @description Count the input tokens of a Responses API request without calling the model.
*
* Follows the OpenAI Responses API spec: https://platform.openai.com/docs/api-reference/responses/input-tokens
*
* ```bash
* curl -X POST http://localhost:4000/v1/responses/input_tokens -H "Content-Type: application/json" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -d '{
* "model": "gpt-4o",
* "input": "Hello, how are you?"
* }'
* ```
*
* Returns: `{"object": "response.input_tokens", "input_tokens": <count>}`
*/
post: operations["responses_input_tokens_responses_input_tokens_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/responses/{response_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Response
* @description Get a response by ID.
*
* Supports both:
* - Polling IDs (litellm_poll_*): Returns cumulative cached content from background responses
* - Provider response IDs: Passes through to provider API
*
* Follows the OpenAI Responses API spec: https://platform.openai.com/docs/api-reference/responses/get
*
* ```bash
* # Get polling response
* curl -X GET http://localhost:4000/v1/responses/litellm_poll_abc123 -H "Authorization: Bearer $LITELLM_MASTER_KEY"
*
* # Get provider response
* curl -X GET http://localhost:4000/v1/responses/resp_abc123 -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*/
get: operations["get_response_responses__response_id__get"];
put?: never;
post?: never;
/**
* Delete Response
* @description Delete a response by ID.
*
* Supports both:
* - Polling IDs (litellm_poll_*): Deletes from Redis cache
* - Provider response IDs: Passes through to provider API
*
* Follows the OpenAI Responses API spec: https://platform.openai.com/docs/api-reference/responses/delete
*
* ```bash
* curl -X DELETE http://localhost:4000/v1/responses/resp_abc123 -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*/
delete: operations["delete_response_responses__response_id__delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/responses/{response_id}/cancel": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Cancel Response
* @description Cancel a response by ID.
*
* Supports both:
* - Polling IDs (litellm_poll_*): Cancels background response and updates status in Redis
* - Provider response IDs: Passes through to provider API
*
* Follows the OpenAI Responses API spec: https://platform.openai.com/docs/api-reference/responses/cancel
*
* ```bash
* # Cancel polling response
* curl -X POST http://localhost:4000/v1/responses/litellm_poll_abc123/cancel -H "Authorization: Bearer $LITELLM_MASTER_KEY"
*
* # Cancel provider response
* curl -X POST http://localhost:4000/v1/responses/resp_abc123/cancel -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*/
post: operations["cancel_response_responses__response_id__cancel_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/responses/{response_id}/input_items": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Response Input Items
* @description List input items for a response.
*/
get: operations["get_response_input_items_responses__response_id__input_items_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/revoke": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Revoke Endpoint
* @description RFC 7009 revocation for the gateway's refresh tokens (``lite logout``): 200 for a known
* client whatever the token's state, 503 when the shared single-use record cannot be written;
* access tokens expire on their own.
*/
post: operations["revoke_endpoint_revoke_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/robots.txt": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Robots
* @description Block all web crawlers from indexing the proxy server endpoints
* This is useful for ensuring that the API endpoints aren't indexed by search engines
*/
get: operations["get_robots_robots_txt_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/router/fields": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Router Fields
* @description Get router settings field definitions without values.
*
* Returns only the field metadata (type, description, default, options) without
* populating field_value. This is useful for UI components that need to know
* what fields to render, but will get the actual values from a different endpoint.
*
* Returns:
* - fields: List of all configurable router settings with their metadata (type, description, default, options)
* The routing_strategy field includes available options extracted from the Router class
* Note: field_value will be None for all fields
* - routing_strategy_descriptions: Descriptions for each routing strategy option
*/
get: operations["get_router_fields_router_fields_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/router/settings": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Router Settings
* @description Get router configuration and available settings.
*
* Returns:
* - fields: List of all configurable router settings with their metadata (type, description, default, options)
* The routing_strategy field includes available options extracted from the Router class
* - current_values: Current values of router settings from config
*/
get: operations["get_router_settings_router_settings_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/routes": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Routes
* @description Get a list of available routes in the FastAPI application.
*/
get: operations["get_routes_routes_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/schedule/anthropic_beta_headers_reload": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Schedule Anthropic Beta Headers Reload
* @description ADMIN ONLY / MASTER KEY Only Endpoint
*
* Schedule periodic reload of the Anthropic beta headers configuration.
* This will create a background job that reloads the configuration every specified hours.
*/
post: operations["schedule_anthropic_beta_headers_reload_schedule_anthropic_beta_headers_reload_post"];
/**
* Cancel Anthropic Beta Headers Reload
* @description ADMIN ONLY / MASTER KEY Only Endpoint
*
* Cancel the scheduled periodic reload of the Anthropic beta headers configuration.
*/
delete: operations["cancel_anthropic_beta_headers_reload_schedule_anthropic_beta_headers_reload_delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/schedule/anthropic_beta_headers_reload/status": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Anthropic Beta Headers Reload Status
* @description ADMIN ONLY / MASTER KEY Only Endpoint
*
* Get the status of the scheduled Anthropic beta headers reload job.
*/
get: operations["get_anthropic_beta_headers_reload_status_schedule_anthropic_beta_headers_reload_status_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/schedule/model_cost_map_reload": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Schedule Model Cost Map Reload
* @description ADMIN ONLY / MASTER KEY Only Endpoint
*
* Schedule periodic reload of the model cost map.
* This will create a background job that reloads the model cost map every specified hours.
*/
post: operations["schedule_model_cost_map_reload_schedule_model_cost_map_reload_post"];
/**
* Cancel Model Cost Map Reload
* @description ADMIN ONLY / MASTER KEY Only Endpoint
*
* Cancel the scheduled periodic reload of the model cost map.
*/
delete: operations["cancel_model_cost_map_reload_schedule_model_cost_map_reload_delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/schedule/model_cost_map_reload/status": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Model Cost Map Reload Status
* @description ADMIN ONLY / MASTER KEY Only Endpoint
*
* Get the status of the scheduled model cost map reload job.
*/
get: operations["get_model_cost_map_reload_status_schedule_model_cost_map_reload_status_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/scim/v2": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Scim Base
* @description Base SCIM v2 endpoint for resource discovery per RFC 7644 Section 4.
*
* Returns a ListResponse of ResourceTypes supported by this SCIM service provider.
* Identity providers (Okta, Azure AD, etc.) use this endpoint for resource discovery.
*/
get: operations["get_scim_base_scim_v2_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/scim/v2/Groups": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Groups
* @description Get a list of groups according to SCIM v2 protocol
*/
get: operations["get_groups_scim_v2_Groups_get"];
put?: never;
/**
* Create Group
* @description Create a group according to SCIM v2 protocol
*/
post: operations["create_group_scim_v2_Groups_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/scim/v2/Groups/{group_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Group
* @description Get a single group by ID according to SCIM v2 protocol
*/
get: operations["get_group_scim_v2_Groups__group_id__get"];
/**
* Update Group
* @description Update a group according to SCIM v2 protocol
*/
put: operations["update_group_scim_v2_Groups__group_id__put"];
post?: never;
/**
* Delete Group
* @description Delete a group according to SCIM v2 protocol
*/
delete: operations["delete_group_scim_v2_Groups__group_id__delete"];
options?: never;
head?: never;
/**
* Patch Group
* @description Patch a group according to SCIM v2 protocol
*/
patch: operations["patch_group_scim_v2_Groups__group_id__patch"];
trace?: never;
};
"/scim/v2/ResourceTypes": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Resource Types
* @description SCIM ResourceTypes endpoint per RFC 7644 Section 4.
*
* Returns a ListResponse of all resource types supported by this service provider.
*/
get: operations["get_resource_types_scim_v2_ResourceTypes_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/scim/v2/ResourceTypes/{resource_type_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Resource Type
* @description Get a single ResourceType by ID per RFC 7644.
*/
get: operations["get_resource_type_scim_v2_ResourceTypes__resource_type_id__get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/scim/v2/Schemas": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Schemas
* @description SCIM Schemas endpoint per RFC 7643 Section 7.
*
* Returns a ListResponse of all schemas supported by this service provider.
*/
get: operations["get_schemas_scim_v2_Schemas_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/scim/v2/Schemas/{schema_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Schema
* @description Get a single Schema by its URI per RFC 7643 Section 7.
*/
get: operations["get_schema_scim_v2_Schemas__schema_id__get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/scim/v2/ServiceProviderConfig": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Service Provider Config
* @description Return SCIM Service Provider Configuration.
*/
get: operations["get_service_provider_config_scim_v2_ServiceProviderConfig_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/scim/v2/Users": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Users
* @description Get a list of users according to SCIM v2 protocol
*/
get: operations["get_users_scim_v2_Users_get"];
put?: never;
/**
* Create User
* @description Create a user according to SCIM v2 protocol
*/
post: operations["create_user_scim_v2_Users_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/scim/v2/Users/{user_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get User
* @description Get a single user by ID according to SCIM v2 protocol
*/
get: operations["get_user_scim_v2_Users__user_id__get"];
/**
* Update User
* @description Update a user according to SCIM v2 protocol (full replacement)
*/
put: operations["update_user_scim_v2_Users__user_id__put"];
post?: never;
/**
* Delete User
* @description Delete a user according to SCIM v2 protocol
*/
delete: operations["delete_user_scim_v2_Users__user_id__delete"];
options?: never;
head?: never;
/**
* Patch User
* @description Patch a user according to SCIM v2 protocol
*/
patch: operations["patch_user_scim_v2_Users__user_id__patch"];
trace?: never;
};
"/scim/v2/placeholders": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List Placeholders
* @description List user rows whose id is another account's SSO identity or email.
*
* An earlier release provisioned a group member it could not match as a user keyed
* by the raw member value, and that row now shadows the account the value really
* names, so every push of that member is refused. This lists those rows so an
* operator can fold each one into the account it shadows with
* ``POST /scim/v2/placeholders/{user_id}/merge``. A row that has an SSO identity of
* its own or owns virtual keys is left out: someone uses that account.
*/
get: operations["list_placeholders_scim_v2_placeholders_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/scim/v2/placeholders/{user_id}/merge": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Merge Placeholder
* @description Fold a placeholder user into the one account its id names by SSO identity or email.
*
* The account is added to every team the placeholder is on, then the placeholder is
* deleted the way ``DELETE /scim/v2/Users/{id}`` deletes a user, so the next group
* push resolves the member value to the real account. Refused with 409 when the row
* has an SSO identity of its own, owns virtual keys, or names no account or several.
*/
post: operations["merge_placeholder_scim_v2_placeholders__user_id__merge_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/search": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Search
* @description Search endpoint for performing web searches.
*
* Follows the Perplexity Search API spec:
* https://docs.perplexity.ai/api-reference/search-post
*
* The search_tool_name can be passed either:
* 1. In the URL path: /v1/search/{search_tool_name}
* 2. In the request body: {"search_tool_name": "..."}
*
* Example with search_tool_name in URL (recommended - keeps body Perplexity-compatible):
* ```bash
* curl -X POST "http://localhost:4000/v1/search/litellm-search" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -H "Content-Type: application/json" -d '{
* "query": "latest AI developments 2024",
* "max_results": 5,
* "search_domain_filter": ["arxiv.org", "nature.com"],
* "country": "US"
* }'
* ```
*
* Example with search_tool_name in body:
* ```bash
* curl -X POST "http://localhost:4000/v1/search" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -H "Content-Type: application/json" -d '{
* "search_tool_name": "litellm-search",
* "query": "latest AI developments 2024",
* "max_results": 5,
* "search_domain_filter": ["arxiv.org", "nature.com"],
* "country": "US"
* }'
* ```
*
* Request Body Parameters (when search_tool_name not in URL):
* - search_tool_name (str, required if not in URL): Name of the search tool configured in router
* - query (str or list[str], required): Search query
* - max_results (int, optional): Maximum number of results (1-20), default 10
* - search_domain_filter (list[str], optional): List of domains to filter (max 20)
* - max_tokens_per_page (int, optional): Max tokens per page, default 1024
* - country (str, optional): Country code filter (e.g., 'US', 'GB', 'DE')
*
* When using URL path parameter, only Perplexity-compatible parameters are needed in body:
* - query (str or list[str], required): Search query
* - max_results (int, optional): Maximum number of results (1-20), default 10
* - search_domain_filter (list[str], optional): List of domains to filter (max 20)
* - max_tokens_per_page (int, optional): Max tokens per page, default 1024
* - country (str, optional): Country code filter (e.g., 'US', 'GB', 'DE')
*
* Response follows Perplexity Search API format:
* ```json
* {
* "object": "search",
* "results": [
* {
* "title": "Result title",
* "url": "https://example.com",
* "snippet": "Result snippet...",
* "date": "2024-01-01",
* "last_updated": "2024-01-01"
* }
* ]
* }
* ```
*/
post: operations["search_search_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/search/tools": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List Search Tools
* @description List all available search tools configured in the router.
*
* This endpoint returns the search tools that are currently loaded and available
* for use with the /v1/search endpoint.
*
* Example:
* ```bash
* curl -X GET "http://localhost:4000/v1/search/tools" -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*
* Response:
* ```json
* {
* "object": "list",
* "data": [
* {
* "search_tool_name": "litellm-search",
* "search_provider": "perplexity",
* "description": "Perplexity search tool"
* }
* ]
* }
* ```
*/
get: operations["list_search_tools_search_tools_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/search/{search_tool_name}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Search
* @description Search endpoint for performing web searches.
*
* Follows the Perplexity Search API spec:
* https://docs.perplexity.ai/api-reference/search-post
*
* The search_tool_name can be passed either:
* 1. In the URL path: /v1/search/{search_tool_name}
* 2. In the request body: {"search_tool_name": "..."}
*
* Example with search_tool_name in URL (recommended - keeps body Perplexity-compatible):
* ```bash
* curl -X POST "http://localhost:4000/v1/search/litellm-search" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -H "Content-Type: application/json" -d '{
* "query": "latest AI developments 2024",
* "max_results": 5,
* "search_domain_filter": ["arxiv.org", "nature.com"],
* "country": "US"
* }'
* ```
*
* Example with search_tool_name in body:
* ```bash
* curl -X POST "http://localhost:4000/v1/search" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -H "Content-Type: application/json" -d '{
* "search_tool_name": "litellm-search",
* "query": "latest AI developments 2024",
* "max_results": 5,
* "search_domain_filter": ["arxiv.org", "nature.com"],
* "country": "US"
* }'
* ```
*
* Request Body Parameters (when search_tool_name not in URL):
* - search_tool_name (str, required if not in URL): Name of the search tool configured in router
* - query (str or list[str], required): Search query
* - max_results (int, optional): Maximum number of results (1-20), default 10
* - search_domain_filter (list[str], optional): List of domains to filter (max 20)
* - max_tokens_per_page (int, optional): Max tokens per page, default 1024
* - country (str, optional): Country code filter (e.g., 'US', 'GB', 'DE')
*
* When using URL path parameter, only Perplexity-compatible parameters are needed in body:
* - query (str or list[str], required): Search query
* - max_results (int, optional): Maximum number of results (1-20), default 10
* - search_domain_filter (list[str], optional): List of domains to filter (max 20)
* - max_tokens_per_page (int, optional): Max tokens per page, default 1024
* - country (str, optional): Country code filter (e.g., 'US', 'GB', 'DE')
*
* Response follows Perplexity Search API format:
* ```json
* {
* "object": "search",
* "results": [
* {
* "title": "Result title",
* "url": "https://example.com",
* "snippet": "Result snippet...",
* "date": "2024-01-01",
* "last_updated": "2024-01-01"
* }
* ]
* }
* ```
*/
post: operations["search_search__search_tool_name__post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/search_tools": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Create Search Tool
* @description Create a new search tool.
*
* Example Request:
* ```bash
* curl -X POST "http://localhost:4000/search_tools" \
* -H "Authorization: Bearer <your_api_key>" \
* -H "Content-Type: application/json" \
* -d '{
* "search_tool": {
* "search_tool_name": "litellm-search",
* "litellm_params": {
* "search_provider": "perplexity",
* "api_key": "sk-..."
* },
* "search_tool_info": {
* "description": "Perplexity search tool"
* }
* }
* }'
* ```
*
* Example Response:
* ```json
* {
* "search_tool_id": "123e4567-e89b-12d3-a456-426614174000",
* "search_tool_name": "litellm-search",
* "litellm_params": {
* "search_provider": "perplexity",
* "api_key": "sk-..."
* },
* "search_tool_info": {
* "description": "Perplexity search tool"
* },
* "created_at": "2023-11-09T12:34:56.789Z",
* "updated_at": "2023-11-09T12:34:56.789Z"
* }
* ```
*/
post: operations["create_search_tool_search_tools_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/search_tools/list": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List Search Tools
* @description List all search tools that are available in the database and config file.
*
* Example Request:
* ```bash
* curl -X GET "http://localhost:4000/search_tools/list" -H "Authorization: Bearer <your_api_key>"
* ```
*
* Example Response:
* ```json
* {
* "search_tools": [
* {
* "search_tool_id": "123e4567-e89b-12d3-a456-426614174000",
* "search_tool_name": "litellm-search",
* "litellm_params": {
* "search_provider": "perplexity",
* "api_key": "sk-***",
* "api_base": "https://api.perplexity.ai"
* },
* "search_tool_info": {
* "description": "Perplexity search tool"
* },
* "created_at": "2023-11-09T12:34:56.789Z",
* "updated_at": "2023-11-09T12:34:56.789Z",
* "is_from_config": false
* },
* {
* "search_tool_name": "config-search-tool",
* "litellm_params": {
* "search_provider": "tavily",
* "api_key": "tvly-***"
* },
* "is_from_config": true
* }
* ]
* }
* ```
*/
get: operations["list_search_tools_search_tools_list_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/search_tools/test_connection": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Test Search Tool Connection
* @description Test connection to a search provider with the given configuration.
*
* Makes a simple test search query to verify the API key and configuration are valid.
*
* Example Request:
* ```bash
* curl -X POST "http://localhost:4000/search_tools/test_connection" \
* -H "Authorization: Bearer <your_api_key>" \
* -H "Content-Type: application/json" \
* -d '{
* "litellm_params": {
* "search_provider": "perplexity",
* "api_key": "sk-..."
* }
* }'
* ```
*
* Example Response (Success):
* ```json
* {
* "status": "success",
* "message": "Successfully connected to perplexity search provider",
* "test_query": "test",
* "results_count": 5
* }
* ```
*
* Example Response (Failure):
* ```json
* {
* "status": "error",
* "message": "Authentication failed: Invalid API key",
* "error_type": "AuthenticationError"
* }
* ```
*/
post: operations["test_search_tool_connection_search_tools_test_connection_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/search_tools/ui/available_providers": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Available Search Providers
* @description Get the list of available search providers with their configuration fields.
*
* Auto-discovers search providers and their UI-friendly names from transformation configs.
*
* Example Request:
* ```bash
* curl -X GET "http://localhost:4000/search_tools/ui/available_providers" \
* -H "Authorization: Bearer <your_api_key>"
* ```
*
* Example Response:
* ```json
* {
* "providers": [
* {
* "provider_name": "perplexity",
* "ui_friendly_name": "Perplexity"
* },
* {
* "provider_name": "tavily",
* "ui_friendly_name": "Tavily"
* }
* ]
* }
* ```
*/
get: operations["get_available_search_providers_search_tools_ui_available_providers_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/search_tools/{search_tool_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Search Tool Info
* @description Get detailed information about a specific search tool by ID.
*
* Example Request:
* ```bash
* curl -X GET "http://localhost:4000/search_tools/123e4567-e89b-12d3-a456-426614174000" \
* -H "Authorization: Bearer <your_api_key>"
* ```
*
* Example Response:
* ```json
* {
* "search_tool_id": "123e4567-e89b-12d3-a456-426614174000",
* "search_tool_name": "litellm-search",
* "litellm_params": {
* "search_provider": "perplexity",
* "api_key": "sk-***"
* },
* "search_tool_info": {
* "description": "Perplexity search tool"
* },
* "created_at": "2023-11-09T12:34:56.789Z",
* "updated_at": "2023-11-09T12:34:56.789Z"
* }
* ```
*/
get: operations["get_search_tool_info_search_tools__search_tool_id__get"];
/**
* Update Search Tool
* @description Update an existing search tool.
*
* Example Request:
* ```bash
* curl -X PUT "http://localhost:4000/search_tools/123e4567-e89b-12d3-a456-426614174000" \
* -H "Authorization: Bearer <your_api_key>" \
* -H "Content-Type: application/json" \
* -d '{
* "search_tool": {
* "search_tool_name": "updated-search",
* "litellm_params": {
* "search_provider": "perplexity",
* "api_key": "sk-new-key"
* },
* "search_tool_info": {
* "description": "Updated search tool"
* }
* }
* }'
* ```
*
* Example Response:
* ```json
* {
* "search_tool_id": "123e4567-e89b-12d3-a456-426614174000",
* "search_tool_name": "updated-search",
* "litellm_params": {
* "search_provider": "perplexity",
* "api_key": "sk-new-key"
* },
* "search_tool_info": {
* "description": "Updated search tool"
* },
* "created_at": "2023-11-09T12:34:56.789Z",
* "updated_at": "2023-11-09T13:45:12.345Z"
* }
* ```
*/
put: operations["update_search_tool_search_tools__search_tool_id__put"];
post?: never;
/**
* Delete Search Tool
* @description Delete a search tool.
*
* Example Request:
* ```bash
* curl -X DELETE "http://localhost:4000/search_tools/123e4567-e89b-12d3-a456-426614174000" \
* -H "Authorization: Bearer <your_api_key>"
* ```
*
* Example Response:
* ```json
* {
* "message": "Search tool 123e4567-e89b-12d3-a456-426614174000 deleted successfully",
* "search_tool_name": "litellm-search"
* }
* ```
*/
delete: operations["delete_search_tool_search_tools__search_tool_id__delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/session/logout": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Session Logout
* @description Revoke the UI session key this request authenticated with.
*
* Only accepts UI session keys (minted by dashboard login); any other
* credential is refused, so this can never be used to delete arbitrary keys.
* Revokes only the presented session, not the user's other sessions.
* Idempotent: logging out an already-revoked session succeeds.
*/
post: operations["session_logout_session_logout_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/settings": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Active Callbacks
* @description Returns a list of litellm level settings
*
* This is useful for debugging and ensuring the proxy server is configured correctly.
*
* Response schema:
* ```
* {
* "alerting": _alerting,
* "litellm.callbacks": litellm_callbacks,
* "litellm.input_callback": litellm_input_callbacks,
* "litellm.failure_callback": litellm_failure_callbacks,
* "litellm.success_callback": litellm_success_callbacks,
* "litellm._async_success_callback": litellm_async_success_callbacks,
* "litellm._async_failure_callback": litellm_async_failure_callbacks,
* "litellm._async_input_callback": litellm_async_input_callbacks,
* "all_litellm_callbacks": all_litellm_callbacks,
* "num_callbacks": len(all_litellm_callbacks),
* "num_alerting": _num_alerting,
* "litellm.request_timeout": litellm.request_timeout,
* }
* ```
*/
get: operations["active_callbacks_settings_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/spend/calculate": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Calculate Spend
* @description Accepts all the params of completion_cost.
*
* Calculate spend **before** making call:
*
* Note: If you see a spend of $0.0 you need to set custom_pricing for your model: https://docs.litellm.ai/docs/proxy/custom_pricing
*
* ```
* curl --location 'http://localhost:4000/spend/calculate'
* --header "Authorization: Bearer $LITELLM_MASTER_KEY"
* --header 'Content-Type: application/json'
* --data '{
* "model": "anthropic.claude-v2",
* "messages": [{"role": "user", "content": "Hey, how'''s it going?"}]
* }'
* ```
*
* Calculate spend **after** making call:
*
* ```
* curl --location 'http://localhost:4000/spend/calculate'
* --header "Authorization: Bearer $LITELLM_MASTER_KEY"
* --header 'Content-Type: application/json'
* --data '{
* "completion_response": {
* "id": "chatcmpl-123",
* "object": "chat.completion",
* "created": 1677652288,
* "model": "gpt-3.5-turbo-0125",
* "system_fingerprint": "fp_44709d6fcb",
* "choices": [{
* "index": 0,
* "message": {
* "role": "assistant",
* "content": "Hello there, how may I assist you today?"
* },
* "logprobs": null,
* "finish_reason": "stop"
* }]
* "usage": {
* "prompt_tokens": 9,
* "completion_tokens": 12,
* "total_tokens": 21
* }
* }
* }'
* ```
*/
post: operations["calculate_spend_spend_calculate_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/spend/capture_rate": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Spend Capture Rate
* @description Compare the spend LiteLLM captured for a provider against that provider's own bill, per UTC day.
*
* Admin only. Reads the provider's billing API with the billing credential set on the proxy
* (OpenAI: `OPENAI_ADMIN_KEY`) and sums `LiteLLM_DailyUserSpend` for the same days.
*
* Example:
* ```
* curl -H "Authorization: Bearer $LITELLM_MASTER_KEY" "http://localhost:4000/spend/capture_rate?provider=openai&start_date=2026-09-17&end_date=2026-09-23"
* ```
*/
get: operations["get_spend_capture_rate_spend_capture_rate_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/spend/keys": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Spend Key Fn
* @description View keys created, ordered by spend.
*
* - Admin callers (PROXY_ADMIN / PROXY_ADMIN_VIEW_ONLY) see every key in
* the database.
* - All other callers (INTERNAL_USER / INTERNAL_USER_VIEW_ONLY, etc.) are
* scoped to keys they own (``user_id == caller``). A caller with no
* ``user_id`` has no scope and receives an empty list rather than the
* full table.
*
* Example Request:
* ```
* curl -X GET "http://0.0.0.0:8000/spend/keys" -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*/
get: operations["spend_key_fn_spend_keys_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/spend/logs": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* View Spend Logs
* @description [DEPRECATED] This endpoint is not paginated and can cause performance issues.
* Please use `/spend/logs/v2` instead for paginated access to spend logs.
*
* Row results are capped at 10,000 most recent entries per response.
*
* View all spend logs, if request_id is provided, only logs for that request_id will be returned
*
* When start_date and end_date are provided:
* - summarize=true (default): Returns aggregated spend data grouped by date (maintains backward compatibility)
* - summarize=false: Returns filtered individual log entries within the date range
*
* Example Request for all logs
* ```
* curl -X GET "http://0.0.0.0:8000/spend/logs" -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*
* Example Request for specific request_id
* ```
* curl -X GET "http://0.0.0.0:8000/spend/logs?request_id=chatcmpl-6dcb2540-d3d7-4e49-bb27-291f863f112e" -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*
* Example Request for specific api_key
* ```
* curl -X GET "http://0.0.0.0:8000/spend/logs?api_key=d5345c0ecc68ae6295c69f91926b2bd379e25481a40c34b5884d157a9f65d8fa" -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*
* Example Request for specific user_id
* ```
* curl -X GET "http://0.0.0.0:8000/spend/[email protected]" -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*
* Example Request for date range with individual logs (unsummarized)
* ```
* curl -X GET "http://0.0.0.0:8000/spend/logs?start_date=2024-01-01&end_date=2024-01-02&summarize=false" -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*/
get: operations["view_spend_logs_spend_logs_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/spend/logs/session/ui": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Ui View Session Spend Logs
* @description Get paginated spend logs for a particular session.
*
* Returns:
* {
* "data": List[LiteLLM_SpendLogs],
* "total": int,
* "page": int,
* "page_size": int,
* "total_pages": int,
* }
*/
get: operations["ui_view_session_spend_logs_spend_logs_session_ui_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/spend/logs/ui": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Ui View Spend Logs
* @description View spend logs with pagination support.
* Available at both `/spend/logs/v2` (public API) and `/spend/logs/ui` (internal UI).
*
* Returns paginated response with data, total, page, page_size, and total_pages.
*
* Example:
* ```
* curl -X GET "http://0.0.0.0:8000/spend/logs/v2?start_date=2025-11-25%2000:00:00&end_date=2025-11-26%2023:59:59&page=1&page_size=50" -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*/
get: operations["ui_view_spend_logs_spend_logs_ui_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/spend/logs/ui/{request_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Ui View Request Response For Request Id
* @description View request / response for a specific request_id
*
* - goes through all callbacks, checks if any of them have a @property -> has_request_response_payload
* - if so, it will return the request and response payload
*/
get: operations["ui_view_request_response_for_request_id_spend_logs_ui__request_id__get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/spend/logs/v2": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Ui View Spend Logs
* @description View spend logs with pagination support.
* Available at both `/spend/logs/v2` (public API) and `/spend/logs/ui` (internal UI).
*
* Returns paginated response with data, total, page, page_size, and total_pages.
*
* Example:
* ```
* curl -X GET "http://0.0.0.0:8000/spend/logs/v2?start_date=2025-11-25%2000:00:00&end_date=2025-11-26%2023:59:59&page=1&page_size=50" -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*/
get: operations["ui_view_spend_logs_spend_logs_v2_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/spend/tags": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* View Spend Tags
* @description LiteLLM Enterprise - View Spend Per Request Tag
*
* - Admin callers (PROXY_ADMIN / PROXY_ADMIN_VIEW_ONLY) see tag aggregates
* across every spend log.
* - All other callers see aggregates over their own spend logs and those of
* teams they may view, exactly like ``/spend/logs/ui``.
*
* Example Request:
* ```
* curl -X GET "http://0.0.0.0:8000/spend/tags" -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*
* Spend with Start Date and End Date
* ```
* curl -X GET "http://0.0.0.0:8000/spend/tags?start_date=2022-01-01&end_date=2022-02-01" -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*/
get: operations["view_spend_tags_spend_tags_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/spend/users": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Spend User Fn
* @description View users created, ordered by spend.
*
* - Admin callers (PROXY_ADMIN / PROXY_ADMIN_VIEW_ONLY) see every user, or
* a specific user when ``user_id`` is supplied.
* - All other callers may only read their own row. If they supply a
* ``user_id`` query parameter that does not match their authenticated
* ``user_id`` the request is rejected with HTTP 403; supplying their
* own id (or none at all) returns just their row. A caller with no
* ``user_id`` on their key has no scope and receives an empty list
* rather than the full table.
*
* Example Request:
* ```
* curl -X GET "http://0.0.0.0:8000/spend/users" -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*
* View User Table row for user_id
* ```
* curl -X GET "http://0.0.0.0:8000/spend/users?user_id=1234" -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*/
get: operations["spend_user_fn_spend_users_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/sso/callback": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Auth Callback
* @description Verify login
*/
get: operations["auth_callback_sso_callback_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/sso/cli/complete/{login_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/** Cli Sso Complete */
post: operations["cli_sso_complete_sso_cli_complete__login_id__post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/sso/cli/poll/{key_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Cli Poll Key
* @description CLI polling endpoint - retrieves session from cache and generates JWT.
*
* Flow:
* 1. First poll (no team_id): Returns teams list without generating JWT
* 2. Second poll (with team_id): Generates JWT with selected team and deletes session
*
* Args:
* key_id: The CLI login session ID
* team_id: Optional team ID to assign to the JWT. If provided, must be one of user's teams.
*/
get: operations["cli_poll_key_sso_cli_poll__key_id__get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/sso/cli/start": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/** Cli Sso Start */
post: operations["cli_sso_start_sso_cli_start_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/sso/debug/callback": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Debug Sso Callback
* @description Returns the OpenID object returned by the SSO provider
*/
get: operations["debug_sso_callback_sso_debug_callback_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/sso/debug/login": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Debug Sso Login
* @description Create Proxy API Keys using Google Workspace SSO. Requires setting PROXY_BASE_URL in .env
* PROXY_BASE_URL should be the your deployed proxy endpoint, e.g. PROXY_BASE_URL="https://litellm-production-7002.up.railway.app/"
* Example:
*/
get: operations["debug_sso_login_sso_debug_login_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/sso/get/ui_settings": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get Ui Settings */
get: operations["get_ui_settings_sso_get_ui_settings_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/sso/key/generate": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Google Login
* @description Create Proxy API Keys using Google Workspace SSO. Requires setting PROXY_BASE_URL in .env
* PROXY_BASE_URL should be the your deployed proxy endpoint, e.g. PROXY_BASE_URL="https://litellm-production-7002.up.railway.app/"
* Example:
*/
get: operations["google_login_sso_key_generate_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/sso/readiness": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Sso Readiness
* @description Health endpoint for checking SSO readiness.
* Checks if the configured SSO provider has all required environment variables set in memory.
*/
get: operations["sso_readiness_sso_readiness_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/sso/saml/callback": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Saml Callback
* @description Assertion Consumer Service. Validates the IdP assertion and issues a UI session.
*/
post: operations["saml_callback_sso_saml_callback_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/sso/saml/login": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Saml Login
* @description SP-initiated SAML login. Redirects the user to the configured IdP.
*/
get: operations["saml_login_sso_saml_login_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/sso/saml/metadata": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Saml Metadata
* @description Service Provider metadata XML, for registering this proxy at the IdP.
*/
get: operations["saml_metadata_sso_saml_metadata_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/systemone": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Ask named questions about a state (System One format)
* @description Judge a `state` against 1 to 128 named `questions` and get one calibrated answer per question name.
*
* Question types are `noul` (probability the answer is yes), `choice` (one label out of `criteria`, with
* `confidence` and `probabilities`), `score` (an index into the `criteria` levels, with `confidence`,
* `legend` and `probabilities`) and `box` (the image region the `instructions` describe, with `confidence`).
* TypeSafe models also take images in `assets`, each one referenced by its key in `state`. Requests that break
* these rules are rejected with a 400 before any provider is called. `model` is any decision model in the proxy model list. OpenAI decision models accept this format
* too, LiteLLM translates it and keeps the answers keyed by question name. Streaming is not supported.
*
* [Docs](https://docs.litellm.ai/docs/decisions)
*/
post: operations["systemone_systemone_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/tag/daily/activity": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Tag Daily Activity
* @description Get daily activity for specific tags or all tags.
*
* Args:
* tags (Optional[str]): Comma-separated list of tags to filter by. If not provided, returns data for all tags.
* start_date (Optional[str]): Start date for the activity period (YYYY-MM-DD).
* end_date (Optional[str]): End date for the activity period (YYYY-MM-DD).
* model (Optional[str]): Filter by model name.
* api_key (Optional[str]): Filter by API key.
* page (int): Page number for pagination.
* page_size (int): Number of items per page.
*
* Returns:
* SpendAnalyticsPaginatedResponse: Paginated response containing daily activity data.
*/
get: operations["get_tag_daily_activity_tag_daily_activity_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/tag/daily/activity/aggregated": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get Tag Daily Activity Aggregated */
get: operations["get_tag_daily_activity_aggregated_tag_daily_activity_aggregated_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/tag/daily/activity/aggregated/keys": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get Tag Daily Activity Aggregated Keys */
get: operations["get_tag_daily_activity_aggregated_keys_tag_daily_activity_aggregated_keys_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/tag/daily/activity/aggregated/model_top_keys": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get Tag Daily Activity Model Top Keys */
get: operations["get_tag_daily_activity_model_top_keys_tag_daily_activity_aggregated_model_top_keys_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/tag/daily/activity/aggregated/search": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get Tag Daily Activity Aggregated Search */
get: operations["get_tag_daily_activity_aggregated_search_tag_daily_activity_aggregated_search_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/tag/daily/activity/export": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get Tag Daily Activity Export */
get: operations["get_tag_daily_activity_export_tag_daily_activity_export_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/tag/dau": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Daily Active Users
* @description Get Daily Active Users (DAU) by tags for the last {MAX_DAYS} days ending on UTC today + 1 day.
*
* This endpoint efficiently calculates unique users per tag for each of the last {MAX_DAYS} days
* using a single optimized SQL query, perfect for dashboard time series visualization.
*
* Args:
* tag_filter: Optional filter to specific tag (legacy)
* tag_filters: Optional filter to multiple specific tags (takes precedence over tag_filter)
*
* Returns:
* ActiveUsersAnalyticsResponse: DAU data by tag for each of the last {MAX_DAYS} days
*/
get: operations["get_daily_active_users_tag_dau_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/tag/delete": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Delete Tag
* @description Delete a tag.
*
* Parameters:
* - name: str - The name of the tag to delete
*/
post: operations["delete_tag_tag_delete_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/tag/distinct": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Distinct User Agent Tags
* @description Get all distinct user agent tags up to a maximum of {MAX_TAGS} tags.
*
* This endpoint returns all unique user agent tags found in the database,
* sorted by frequency of usage.
*
* Returns:
* DistinctTagsResponse: List of distinct user agent tags
*/
get: operations["get_distinct_user_agent_tags_tag_distinct_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/tag/info": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Info Tag
* @description Get information about specific tags.
*
* Parameters:
* - names: List[str] - List of tag names to get information for
*/
post: operations["info_tag_tag_info_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/tag/list": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List Tags
* @description List all available tags with their budget information.
*/
get: operations["list_tags_tag_list_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/tag/mau": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Monthly Active Users
* @description Get Monthly Active Users (MAU) by tags for the last {MAX_MONTHS} months ending on UTC today + 1 day.
*
* Shows month-by-month breakdown:
* - Month 1 (Nov): Earliest month (7 months ago, 30-day period)
* - Month 2 (Dec): Next month (6 months ago)
* - Month 3 (Jan): Next month (5 months ago)
* - ... and so on for {MAX_MONTHS} months total
* - Month 7: Most recent month ending on UTC today + 1 day
*
* Args:
* tag_filter: Optional filter to specific tag (legacy)
* tag_filters: Optional filter to multiple specific tags (takes precedence over tag_filter)
*
* Returns:
* ActiveUsersAnalyticsResponse: MAU data by tag for each of the last {MAX_MONTHS} months with descriptive month labels (e.g., "Month 1 (Nov)")
*/
get: operations["get_monthly_active_users_tag_mau_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/tag/new": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* New Tag
* @description Create a new tag.
*
* Parameters:
* - name: str - The name of the tag
* - description: Optional[str] - Description of what this tag represents
* - models: List[str] - List of either 'model_id' or 'model_name' allowed for this tag
* - budget_id: Optional[str] - The id for a budget (tpm/rpm/max budget) for the tag
*
* ### IF NO BUDGET ID - CREATE ONE WITH THESE PARAMS ###
* - max_budget: Optional[float] - Max budget for tag
* - tpm_limit: Optional[int] - Max tpm limit for tag
* - rpm_limit: Optional[int] - Max rpm limit for tag
* - max_parallel_requests: Optional[int] - Max parallel requests for tag
* - soft_budget: Optional[float] - Get a slack alert when this soft budget is reached
* - model_max_budget: Optional[dict] - Max budget for a specific model
* - budget_duration: Optional[str] - Frequency of resetting tag budget
*/
post: operations["new_tag_tag_new_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/tag/summary": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Tag Summary
* @description Get summary analytics for tags including unique users, requests, tokens, and spend.
*
* Args:
* start_date: Start date for the analytics period (YYYY-MM-DD)
* end_date: End date for the analytics period (YYYY-MM-DD)
* tag_filter: Optional filter to specific tag (legacy)
* tag_filters: Optional filter to multiple specific tags (takes precedence over tag_filter)
*
* Returns:
* TagSummaryResponse: Summary analytics data by tag
*/
get: operations["get_tag_summary_tag_summary_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/tag/update": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Update Tag
* @description Update an existing tag.
*
* Parameters:
* - name: str - The name of the tag to update
* - description: Optional[str] - Updated description
* - models: List[str] - Updated list of allowed LLM models
* - budget_id: Optional[str] - The id for a budget to associate with the tag
*
* ### BUDGET UPDATE PARAMS ###
* - max_budget: Optional[float] - Max budget for tag
* - tpm_limit: Optional[int] - Max tpm limit for tag
* - rpm_limit: Optional[int] - Max rpm limit for tag
* - max_parallel_requests: Optional[int] - Max parallel requests for tag
* - soft_budget: Optional[float] - Get a slack alert when this soft budget is reached
* - model_max_budget: Optional[dict] - Max budget for a specific model
* - budget_duration: Optional[str] - Frequency of resetting tag budget
*/
post: operations["update_tag_tag_update_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/tag/user-agent/per-user-analytics": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Per User Analytics
* @description Get per-user analytics including successful requests, tokens, and spend by individual users.
*
* This endpoint provides usage metrics broken down by individual users based on their
* tag activity during the last 30 days ending on UTC today + 1 day.
*
* Args:
* tag_filter: Optional filter to specific tag (legacy)
* tag_filters: Optional filter to multiple specific tags (takes precedence over tag_filter)
* page: Page number for pagination
* page_size: Number of items per page
*
* Returns:
* PerUserAnalyticsResponse: Analytics data broken down by individual users for the last 30 days
*/
get: operations["get_per_user_analytics_tag_user_agent_per_user_analytics_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/tag/wau": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Weekly Active Users
* @description Get Weekly Active Users (WAU) by tags for the last {MAX_WEEKS} weeks ending on UTC today + 1 day.
*
* Shows week-by-week breakdown:
* - Week 1 (Jan 1): Earliest week (7 weeks ago)
* - Week 2 (Jan 8): Next week (6 weeks ago)
* - Week 3 (Jan 15): Next week (5 weeks ago)
* - ... and so on for {MAX_WEEKS} weeks total
* - Week 7: Most recent week ending on UTC today + 1 day
*
* Args:
* tag_filter: Optional filter to specific tag (legacy)
* tag_filters: Optional filter to multiple specific tags (takes precedence over tag_filter)
*
* Returns:
* ActiveUsersAnalyticsResponse: WAU data by tag for each of the last {MAX_WEEKS} weeks with descriptive week labels (e.g., "Week 1 (Jan 1)")
*/
get: operations["get_weekly_active_users_tag_wau_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/team/available": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** List Available Teams */
get: operations["list_available_teams_team_available_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/team/block": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Block Team
* @description Blocks all calls from keys with this team id.
*
* Parameters:
* - team_id: str - Required. The unique identifier of the team to block.
*
* Example:
* ```
* curl --location 'http://0.0.0.0:4000/team/block' --header "Authorization: Bearer $LITELLM_MASTER_KEY" --header 'Content-Type: application/json' --data '{
* "team_id": "team-1234"
* }'
* ```
*
* Returns:
* - The updated team record with blocked=True
*/
post: operations["block_team_team_block_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/team/bulk_member_add": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Bulk Team Member Add
* @description Bulk add multiple members to a team at once.
*
* This endpoint reuses the same logic as /team/member_add but provides a bulk-friendly response format.
*
* Parameters:
* - team_id: str - The ID of the team to add members to
* - members: List[Member] - List of members to add to the team
* - all_users: Optional[bool] - Flag to add all users on Proxy to the team
* - max_budget_in_team: Optional[float] - Maximum budget allocated to each user within the team
*
* Returns:
* - results: List of individual member addition results
* - total_requested: Total number of members requested for addition
* - successful_additions: Number of successful additions
* - failed_additions: Number of failed additions
* - updated_team: The updated team object
*
* Example request:
* ```bash
* curl --location 'http://0.0.0.0:4000/team/bulk_member_add' --header "Authorization: Bearer $LITELLM_MASTER_KEY" --header 'Content-Type: application/json' --data '{
* "team_id": "team-1234",
* "members": [
* {
* "user_id": "user1",
* "role": "user"
* },
* {
* "user_email": "user2@example.com",
* "role": "admin"
* }
* ],
* "max_budget_in_team": 100.0
* }'
* ```
*/
post: operations["bulk_team_member_add_team_bulk_member_add_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/team/daily/activity": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Team Daily Activity
* @description Get daily activity for specific teams or all teams.
*
* Args:
* team_ids (Optional[str]): Comma-separated list of team IDs to filter by. If not provided, returns data for all teams.
* start_date (Optional[str]): Start date for the activity period (YYYY-MM-DD).
* end_date (Optional[str]): End date for the activity period (YYYY-MM-DD).
* model (Optional[str]): Filter by model name.
* api_key (Optional[str]): Filter by API key.
* page (int): Page number for pagination.
* page_size (int): Number of items per page.
* exclude_team_ids (Optional[str]): Comma-separated list of team IDs to exclude.
* Returns:
* SpendAnalyticsPaginatedResponse: Paginated response containing daily activity data.
*/
get: operations["get_team_daily_activity_team_daily_activity_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/team/daily/activity/aggregated": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Team Daily Activity Aggregated
* @description Aggregated daily activity for teams without pagination, including per-team breakdown.
*
* One SQL GROUPING SETS pass returns every day in the range regardless of row
* volume, so callers never reassemble pages. Same response shape as the
* paginated endpoint with page metadata pinned to a single page.
*
* Args:
* team_ids (Optional[str]): Comma-separated list of team IDs to filter by. If not provided, returns data for all teams.
* start_date (Optional[str]): Start date for the activity period (YYYY-MM-DD).
* end_date (Optional[str]): End date for the activity period (YYYY-MM-DD).
* model (Optional[str]): Filter by model name.
* api_key (Optional[str]): Filter by API key.
* exclude_team_ids (Optional[str]): Comma-separated list of team IDs to exclude.
* timezone (Optional[int]): Timezone offset in minutes from UTC, matching JavaScript's Date.getTimezoneOffset() convention.
* Returns:
* SpendAnalyticsPaginatedResponse: Response containing all daily activity data for the range.
*/
get: operations["get_team_daily_activity_aggregated_team_daily_activity_aggregated_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/team/daily/activity/aggregated/keys": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get Team Daily Activity Aggregated Keys */
get: operations["get_team_daily_activity_aggregated_keys_team_daily_activity_aggregated_keys_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/team/daily/activity/aggregated/model_top_keys": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get Team Daily Activity Model Top Keys */
get: operations["get_team_daily_activity_model_top_keys_team_daily_activity_aggregated_model_top_keys_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/team/daily/activity/aggregated/search": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get Team Daily Activity Aggregated Search */
get: operations["get_team_daily_activity_aggregated_search_team_daily_activity_aggregated_search_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/team/daily/activity/export": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get Team Daily Activity Export */
get: operations["get_team_daily_activity_export_team_daily_activity_export_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/team/delete": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Delete Team
* @description delete team and associated team keys
*
* Parameters:
* - team_ids: List[str] - Required. List of team IDs to delete. Example: ["team-1234", "team-5678"]
*
* ```
* curl --location 'http://0.0.0.0:4000/team/delete' --header "Authorization: Bearer $LITELLM_MASTER_KEY" --header 'Content-Type: application/json' --data-raw '{
* "team_ids": ["8d916b1c-510d-4894-a334-1c16a93344f5"]
* }'
* ```
*/
post: operations["delete_team_team_delete_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/team/filter/ui": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Ui View Teams
* @description [PROXY-ADMIN ONLY] Filter teams based on partial match of team_id or team_alias with pagination.
*
* Args:
* user_id (Optional[str]): Partial user ID to search for
* user_email (Optional[str]): Partial email to search for
* page (int): Page number for pagination (starts at 1)
* page_size (int): Number of items per page (max 100)
* user_api_key_dict (UserAPIKeyAuth): User authentication information
*
* Returns:
* List[LiteLLM_SpendLogs]: Paginated list of matching user records
*/
get: operations["ui_view_teams_team_filter_ui_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/team/info": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Team Info
* @description get info on team + related keys
*
* Parameters:
* - team_id: str - Required. The unique identifier of the team to get info on.
*
* ```
* curl --location 'http://localhost:4000/team/info?team_id=your_team_id_here' --header 'Authorization: Bearer your_api_key_here'
* ```
*/
get: operations["team_info_team_info_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/team/key/bulk_update": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Bulk Update Team Keys
* @description Apply one update payload to many keys inside a single team.
*
* Pass `team_id` plus either `key_ids` or `all_keys_in_team=True`. The
* `update_fields` payload is broadcast to every selected key. Per-key
* failures are returned in `failed_updates` rather than aborting the batch.
*
* Callable by proxy admins, or by team admins with `KEY_UPDATE` permission.
*/
post: operations["bulk_update_team_keys_team_key_bulk_update_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/team/list": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List Team
* @description ```
* curl --location --request GET 'http://0.0.0.0:4000/team/list' --header "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*
* Parameters:
* - user_id: str - Optional. If passed will only return teams that the user_id is a member of.
* - organization_id: str - Optional. If passed will only return teams that belong to the organization_id. Pass 'default_organization' to get all teams without organization_id.
*/
get: operations["list_team_team_list_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/team/member_add": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Team Member Add
* @description Add new members (either via user_email or user_id) to a team
*
* If user doesn't exist, new user row will also be added to User Table
*
* Only proxy_admin or admin of team, allowed to access this endpoint.
* ```
*
* curl -X POST 'http://0.0.0.0:4000/team/member_add' -H "Authorization: Bearer $LITELLM_MASTER_KEY" -H 'Content-Type: application/json' -d '{"team_id": "45e3e396-ee08-4a61-a88e-16b3ce7e0849", "member": {"role": "user", "user_id": "[email protected]"}}'
*
* ```
*/
post: operations["team_member_add_team_member_add_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/team/member_delete": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Team Member Delete
* @description [BETA]
*
* delete members (either via user_email or user_id) from a team
*
* If user doesn't exist, an exception will be raised
* ```
* curl -X POST 'http://0.0.0.0:8000/team/member_delete'
* -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* -H 'Content-Type: application/json'
* -d '{
* "team_id": "45e3e396-ee08-4a61-a88e-16b3ce7e0849",
* "user_id": "[email protected]"
* }'
* ```
*/
post: operations["team_member_delete_team_member_delete_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/team/member_update": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Team Member Update
* @description [BETA]
*
* Update team member budgets and team member role
*/
post: operations["team_member_update_team_member_update_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/team/metadata_schema": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Team Metadata Schema
* @description Get the team metadata fields declared in ``general_settings.team_metadata_schema``.
*
* The UI uses this to prepopulate the team metadata form with the declared
* keys. Returns an empty ``fields`` list when no schema is configured. This
* schema is advisory; server-side enforcement stays with
* ``custom_team_metadata_validate``.
*/
get: operations["get_team_metadata_schema_team_metadata_schema_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/team/model/add": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Team Model Add
* @description Add models to a team's allowed model list. Only proxy admin or team admin can add models.
*
* Parameters:
* - team_id: str - Required. The team to add models to
* - models: List[str] - Required. List of models to add to the team
*
* Example Request:
* ```
* curl --location 'http://0.0.0.0:4000/team/model/add' --header "Authorization: Bearer $LITELLM_MASTER_KEY" --header 'Content-Type: application/json' --data '{
* "team_id": "team-1234",
* "models": ["gpt-4", "claude-2"]
* }'
* ```
*/
post: operations["team_model_add_team_model_add_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/team/model/delete": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Team Model Delete
* @description Remove models from a team's allowed model list. Only proxy admin or team admin can remove models.
*
* Parameters:
* - team_id: str - Required. The team to remove models from
* - models: List[str] - Required. List of models to remove from the team
*
* Example Request:
* ```
* curl --location 'http://0.0.0.0:4000/team/model/delete' --header "Authorization: Bearer $LITELLM_MASTER_KEY" --header 'Content-Type: application/json' --data '{
* "team_id": "team-1234",
* "models": ["gpt-4"]
* }'
* ```
*/
post: operations["team_model_delete_team_model_delete_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/team/new": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* New Team
* @description Allow users to create a new team. Apply user permissions to their team.
*
* 👉 [Detailed Doc on setting team budgets](https://docs.litellm.ai/docs/proxy/team_budgets)
*
*
* Parameters:
* - team_alias: Optional[str] - User defined team alias
* - team_id: Optional[str] - The team id of the user. If none passed, we'll generate it.
* - members_with_roles: List[{"role": "admin" or "user", "user_id": "<user-id>"}] - A list of users and their roles in the team. Get user_id when making a new user via `/user/new`.
* - team_member_permissions: Optional[List[str]] - A list of routes that non-admin team members can access. example: ["/key/generate", "/key/update", "/key/delete"]
* - metadata: Optional[dict] - Metadata for team, store information for team. Example metadata = {"extra_info": "some info"}
* - model_rpm_limit: Optional[Dict[str, int]] - The RPM (Requests Per Minute) limit for this team - applied across all keys for this team.
* - model_tpm_limit: Optional[Dict[str, int]] - The TPM (Tokens Per Minute) limit for this team - applied across all keys for this team.
* - default_estimated_output_tokens: Optional[int] - Expected output tokens reserved for TPM limiting when a request omits max_tokens, for keys on this team that do not set their own. Positive integer.
* - default_estimated_output_tokens_per_model: Optional[Dict[str, int]] - Per-model override of the above. Example: {"gpt-4": 4096, "gpt-3.5-turbo": 1024}
* - mcp_rpm_limit: Optional[Dict[str, int]] - Per-MCP-server RPM limit for this team, keyed by MCP server name (alias if set, else the configured name). Example: {"github": 100, "slack": 200}. Applied across all keys for this team.
* - tpm_limit: Optional[int] - The TPM (Tokens Per Minute) limit for this team - all keys with this team_id will have at max this TPM limit
* - rpm_limit: Optional[int] - The RPM (Requests Per Minute) limit for this team - all keys associated with this team_id will have at max this RPM limit
* - tpd_limit: Optional[int] - The TPD (Tokens Per Day) limit for this team. Batch submissions are charged against it instead of tpm_limit/rpm_limit
* - rpm_limit_type: Optional[Literal["guaranteed_throughput", "best_effort_throughput"]] - The type of RPM limit enforcement. Use "guaranteed_throughput" to raise an error if overallocating RPM, or "best_effort_throughput" for best effort enforcement.
* - tpm_limit_type: Optional[Literal["guaranteed_throughput", "best_effort_throughput"]] - The type of TPM limit enforcement. Use "guaranteed_throughput" to raise an error if overallocating TPM, or "best_effort_throughput" for best effort enforcement.
* - max_budget: Optional[float] - The maximum budget allocated to the team - all keys for this team_id will have at max this max_budget
* - soft_budget: Optional[float] - The soft budget threshold for the team. If max_budget is set, soft_budget must be strictly lower than max_budget. Can be set independently if max_budget is not set.
* - budget_duration: Optional[str] - The duration of the budget for the team. Doc [here](https://docs.litellm.ai/docs/proxy/team_budgets)
* - models: Optional[list] - A list of models associated with the team - all keys for this team_id will have at most, these models. If empty, assumes all models are allowed.
* - blocked: bool - Flag indicating if the team is blocked or not - will stop all calls from keys with this team_id.
* - members: Optional[List] - Control team members via `/team/member/add` and `/team/member/delete`.
* - tags: Optional[List[str]] - Tags for [tracking spend](https://litellm.vercel.app/docs/proxy/enterprise#tracking-spend-for-custom-tags) and/or doing [tag-based routing](https://litellm.vercel.app/docs/proxy/tag_routing).
* - prompts: Optional[List[str]] - List of prompts that the team is allowed to use.
* - organization_id: Optional[str] - The organization id of the team. Default is None. Create via `/organization/new`.
* - model_aliases: Optional[dict] - Model aliases for the team. [Docs](https://docs.litellm.ai/docs/proxy/team_based_routing#create-team-with-model-alias)
* - model_max_budget: Optional[dict] - Per-model max budget every key on the team inherits unless the key sets its own for that model. Example: {"gpt-4o": {"max_budget": 10, "budget_duration": "1d"}}
* - guardrails: Optional[List[str]] - Guardrails for the team. [Docs](https://docs.litellm.ai/docs/proxy/guardrails)
* - policies: Optional[List[str]] - Policies for the team. [Docs](https://docs.litellm.ai/docs/proxy/guardrails/guardrail_policies)
* - disable_global_guardrails: Optional[bool] - Whether to disable global guardrails for the team. Proxy admin only.
* - require_trace_id: Optional[bool] - Reject LLM, MCP and agent requests from this team that carry no trace ID.
* - object_permission: Optional[LiteLLM_ObjectPermissionBase] - team-specific object permission. Example - {"vector_stores": ["vector_store_1", "vector_store_2"], "agents": ["agent_1", "agent_2"], "agent_access_groups": ["dev_group"]}. IF null or {} then no object permission.
* - team_member_budget: Optional[float] - The maximum budget allocated to an individual team member.
* - team_member_budget_duration: Optional[str] - The duration of the budget for the team member. Doc [here](https://docs.litellm.ai/docs/proxy/team_budgets)
* - team_member_model_max_budget: Optional[dict] - Per-model budgets shared by team members without an override.
* - team_member_rpm_limit: Optional[int] - The RPM (Requests Per Minute) limit for individual team members.
* - team_member_tpm_limit: Optional[int] - The TPM (Tokens Per Minute) limit for individual team members.
* - team_member_key_duration: Optional[str] - The duration for a team member's key. e.g. "1d", "1w", "1mo"
* - allowed_passthrough_routes: Optional[List[str]] - List of allowed pass through routes for the team.
* - denied_passthrough_routes: Optional[List[str]] - List of pass through routes the team's keys may not call, even if allowed. Applies together with each key's `denied_passthrough_routes`.
* - allowed_vector_store_indexes: Optional[List[dict]] - List of allowed vector store indexes for the key. Example - [{"index_name": "my-index", "index_permissions": ["write", "read"]}]. If specified, the key will only be able to use these specific vector store indexes. Create index, using `/v1/indexes` endpoint.
* - secret_manager_settings: Optional[dict] - Secret manager settings for the team. [Docs](https://docs.litellm.ai/docs/secret_managers/overview)
* - router_settings: Optional[UpdateRouterConfig] - team-specific router settings. Example - {"model_group_retry_policy": {"gpt-4": {"RateLimitErrorRetries": 5}}}. IF null or {} then no router settings.
* - access_group_ids: Optional[List[str]] - List of access group IDs to associate with the team. Access groups define which models the team can access. Example - ["access_group_1", "access_group_2"].
* - enforced_file_expires_after: Optional[dict] - Enforced file expiration policy for the team. Keys created under this team will inherit this policy for file uploads. Example - {"anchor": "created_at", "days": 30}.
* - enforced_batch_output_expires_after: Optional[dict] - Enforced batch output file expiration policy for the team. Keys created under this team will inherit this policy for batch output files. Example - {"anchor": "created_at", "days": 30}.
* - budget_limits: Optional[list] - List of concurrent budget windows for the team. Each window specifies a budget_limit, time_period, and optional budget_duration. Example - [{"budget_limit": 10.0, "time_period": "1d"}, {"budget_limit": 50.0, "time_period": "7d"}].
* - default_team_member_models: Optional[List[str]] - Default models assigned to new team members when they join this team. Must be a subset of the team's models.
*
* Returns:
* - team_id: (str) Unique team id - used for tracking spend across multiple keys for same team id.
*
* _deprecated_params:
* - admins: list - A list of user_id's for the admin role
* - users: list - A list of user_id's for the user role
*
* Example Request:
* ```
* curl --location 'http://0.0.0.0:4000/team/new' --header "Authorization: Bearer $LITELLM_MASTER_KEY" --header 'Content-Type: application/json' --data '{
* "team_alias": "my-new-team_2",
* "members_with_roles": [{"role": "admin", "user_id": "user-1234"},
* {"role": "user", "user_id": "user-2434"}]
* }'
*
* ```
*
* ```
* curl --location 'http://0.0.0.0:4000/team/new' --header "Authorization: Bearer $LITELLM_MASTER_KEY" --header 'Content-Type: application/json' --data '{
* "team_alias": "QA Prod Bot",
* "max_budget": 0.000000001,
* "budget_duration": "1d"
* }'
* ```
*/
post: operations["new_team_team_new_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/team/permissions_bulk_update": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Bulk Update Team Member Permissions
* @description Append permissions to existing teams.
*
* Either pass team_ids to target specific teams, or set
* apply_to_all_teams=True to update every team. For each team,
* the provided permissions are merged with the team's existing
* permissions (duplicates are skipped).
*/
post: operations["bulk_update_team_member_permissions_team_permissions_bulk_update_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/team/permissions_list": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Team Member Permissions
* @description Get the team member permissions for a team
*/
get: operations["team_member_permissions_team_permissions_list_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/team/permissions_update": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Update Team Member Permissions
* @description Update the team member permissions for a team
*/
post: operations["update_team_member_permissions_team_permissions_update_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/team/spend/by_user": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Team Spend By User
* @description Spend per user within the given teams, attributed per request from spend logs.
*
* Proxy admins may query any team. Team admins and members holding the
* `/team/daily/activity` permission see every user of the requested teams;
* other members only see their own row.
*/
get: operations["get_team_spend_by_user_team_spend_by_user_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/team/spend/report": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Team Spend Report
* @description Get spend for the calling key's team over a date range, grouped by api_key with a per-model breakdown.
*
* Callable by any key that belongs to a team: non-admin callers are always
* scoped to their key's team_id, while proxy admins may pass `?team_id=` to
* view any team.
*/
get: operations["get_team_spend_report_team_spend_report_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/team/unblock": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Unblock Team
* @description Unblocks a previously blocked team, re-enabling calls from keys with this team id.
*
* Parameters:
* - team_id: str - Required. The unique identifier of the team to unblock.
*
* Example:
* ```
* curl --location 'http://0.0.0.0:4000/team/unblock' --header "Authorization: Bearer $LITELLM_MASTER_KEY" --header 'Content-Type: application/json' --data '{
* "team_id": "team-1234"
* }'
* ```
*/
post: operations["unblock_team_team_unblock_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/team/update": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Update Team
* @description Use `/team/member_add` AND `/team/member/delete` to add/remove new team members
*
* You can now update team budget / rate limits via /team/update
*
* Parameters:
* - team_id: str - The team id of the user. Required param.
* - team_alias: Optional[str] - User defined team alias
* - team_member_permissions: Optional[List[str]] - A list of routes that non-admin team members can access. example: ["/key/generate", "/key/update", "/key/delete"]
* - metadata: Optional[dict] - Metadata for team, store information for team. Example metadata = {"team": "core-infra", "app": "app2", "email": "[email protected]" }
* - tpm_limit: Optional[int] - The TPM (Tokens Per Minute) limit for this team - all keys with this team_id will have at max this TPM limit
* - rpm_limit: Optional[int] - The RPM (Requests Per Minute) limit for this team - all keys associated with this team_id will have at max this RPM limit
* - tpd_limit: Optional[int] - The TPD (Tokens Per Day) limit for this team. Batch submissions are charged against it instead of tpm_limit/rpm_limit
* - max_budget: Optional[float] - The maximum budget allocated to the team - all keys for this team_id will have at max this max_budget
* - soft_budget: Optional[float] - The soft budget threshold for the team. If max_budget is set (either in the request or existing), soft_budget must be strictly lower than max_budget. Can be set independently if max_budget is not set.
* - budget_duration: Optional[str] - The duration of the budget for the team. Doc [here](https://docs.litellm.ai/docs/proxy/team_budgets)
* - models: Optional[list] - A list of models associated with the team - all keys for this team_id will have at most, these models. If empty, assumes all models are allowed.
* - prompts: Optional[List[str]] - List of prompts that the team is allowed to use.
* - blocked: bool - Flag indicating if the team is blocked or not - will stop all calls from keys with this team_id.
* - tags: Optional[List[str]] - Tags for [tracking spend](https://litellm.vercel.app/docs/proxy/enterprise#tracking-spend-for-custom-tags) and/or doing [tag-based routing](https://litellm.vercel.app/docs/proxy/tag_routing).
* - organization_id: Optional[str] - The organization id of the team. Default is None. Create via `/organization/new`.
* - model_aliases: Optional[dict] - Model aliases for the team. [Docs](https://docs.litellm.ai/docs/proxy/team_based_routing#create-team-with-model-alias)
* - model_max_budget: Optional[dict] - Per-model max budget every key on the team inherits unless the key sets its own for that model. Example: {"gpt-4o": {"max_budget": 10, "budget_duration": "1d"}}
* - guardrails: Optional[List[str]] - Guardrails for the team. [Docs](https://docs.litellm.ai/docs/proxy/guardrails)
* - policies: Optional[List[str]] - Policies for the team. [Docs](https://docs.litellm.ai/docs/proxy/guardrails/guardrail_policies)
* - disable_global_guardrails: Optional[bool] - Whether to disable global guardrails for the team. Proxy admin only.
* - require_trace_id: Optional[bool] - Reject LLM, MCP and agent requests from this team that carry no trace ID.
* - object_permission: Optional[LiteLLM_ObjectPermissionBase] - team-specific object permission. Example - {"vector_stores": ["vector_store_1", "vector_store_2"], "agents": ["agent_1", "agent_2"], "agent_access_groups": ["dev_group"]}. IF null or {} then no object permission.
* - team_member_budget: Optional[float] - The maximum budget allocated to an individual team member.
* - team_member_budget_duration: Optional[str] - The duration of the budget for the team member. Doc [here](https://docs.litellm.ai/docs/proxy/team_budgets)
* - team_member_model_max_budget: Optional[dict] - Per-model budgets shared by team members without an override.
* - team_member_rpm_limit: Optional[int] - The RPM (Requests Per Minute) limit for individual team members.
* - team_member_tpm_limit: Optional[int] - The TPM (Tokens Per Minute) limit for individual team members.
* - team_member_key_duration: Optional[str] - The duration for a team member's key. e.g. "1d", "1w", "1mo"
* - allowed_passthrough_routes: Optional[List[str]] - List of allowed pass through routes for the team.
* - denied_passthrough_routes: Optional[List[str]] - List of pass through routes the team's keys may not call, even if allowed. Applies together with each key's `denied_passthrough_routes`.
* - model_rpm_limit: Optional[Dict[str, int]] - The RPM (Requests Per Minute) limit per model for this team. Example: {"gpt-4": 100, "gpt-3.5-turbo": 200}
* - model_tpm_limit: Optional[Dict[str, int]] - The TPM (Tokens Per Minute) limit per model for this team. Example: {"gpt-4": 10000, "gpt-3.5-turbo": 20000}
* - default_estimated_output_tokens: Optional[int] - Expected output tokens reserved for TPM limiting when a request omits max_tokens, for keys on this team that do not set their own. Positive integer.
* - default_estimated_output_tokens_per_model: Optional[Dict[str, int]] - Per-model override of the above. Example: {"gpt-4": 4096, "gpt-3.5-turbo": 1024}
* - mcp_rpm_limit: Optional[Dict[str, int]] - Per-MCP-server RPM limit for this team, keyed by MCP server name (alias if set, else the configured name). Example: {"github": 100, "slack": 200}. Applied across all keys for this team.
* Example - update team TPM Limit
* - allowed_vector_store_indexes: Optional[List[dict]] - List of allowed vector store indexes for the key. Example - [{"index_name": "my-index", "index_permissions": ["write", "read"]}]. If specified, the key will only be able to use these specific vector store indexes. Create index, using `/v1/indexes` endpoint.
* - secret_manager_settings: Optional[dict] - Secret manager settings for the team. [Docs](https://docs.litellm.ai/docs/secret_managers/overview)
* - router_settings: Optional[UpdateRouterConfig] - team-specific router settings. Example - {"model_group_retry_policy": {"gpt-4": {"RateLimitErrorRetries": 5}}}. IF null or {} then no router settings.
* - access_group_ids: Optional[List[str]] - List of access group IDs to associate with the team. Access groups define which models the team can access. Example - ["access_group_1", "access_group_2"].
* - enforced_file_expires_after: Optional[dict] - Enforced file expiration policy for the team. Keys created under this team will inherit this policy for file uploads. Example - {"anchor": "created_at", "days": 30}.
* - enforced_batch_output_expires_after: Optional[dict] - Enforced batch output file expiration policy for the team. Keys created under this team will inherit this policy for batch output files. Example - {"anchor": "created_at", "days": 30}.
* - budget_limits: Optional[list] - List of concurrent budget windows for the team. Each window specifies a budget_limit, time_period, and optional budget_duration. Example - [{"budget_limit": 10.0, "time_period": "1d"}, {"budget_limit": 50.0, "time_period": "7d"}].
* - default_team_member_models: Optional[List[str]] - Default models assigned to new team members when they join this team. Must be a subset of the team's models.
*
* ```
* curl --location 'http://0.0.0.0:4000/team/update' --header "Authorization: Bearer $LITELLM_MASTER_KEY" --header 'Content-Type: application/json' --data-raw '{
* "team_id": "8d916b1c-510d-4894-a334-1c16a93344f5",
* "tpm_limit": 100
* }'
* ```
*
* Example - Update Team `max_budget` budget
* ```
* curl --location 'http://0.0.0.0:4000/team/update' --header "Authorization: Bearer $LITELLM_MASTER_KEY" --header 'Content-Type: application/json' --data-raw '{
* "team_id": "8d916b1c-510d-4894-a334-1c16a93344f5",
* "max_budget": 10
* }'
* ```
*/
post: operations["update_team_team_update_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/team/{team_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
/**
* Patch Team
* @description Partially update a team using RFC 7386 JSON Merge Patch semantics.
*
* `team_id` is taken from the path; a `team_id` in the body is accepted only when it
* matches. `metadata` is merged with the team's stored metadata rather than replacing
* it: an omitted key is preserved, `key: null` deletes it, and any other value
* overwrites (recursing into nested objects). Every other field behaves exactly like
* `POST /team/update` (omitted preserves, a value overwrites). Returns the full
* updated team.
*
* ```
* curl --location --request PATCH 'http://0.0.0.0:4000/team/8d916b1c-510d-4894-a334-1c16a93344f5' --header "Authorization: Bearer $LITELLM_MASTER_KEY" --header 'Content-Type: application/json' --data-raw '{
* "metadata": {"cost_center": "1234", "deprecated_key": null}
* }'
* ```
*/
patch: operations["patch_team_team__team_id__patch"];
trace?: never;
};
"/team/{team_id}/callback": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Team Callbacks
* @description Get the success/failure callbacks and variables for a team
*
* Parameters:
* - team_id (str, required): The unique identifier for the team
*
* Example curl:
* ```
* curl -X GET 'http://localhost:4000/team/dbe2f686-a686-4896-864a-4c3924458709/callback' -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*
* This will return the callback settings for the team with id dbe2f686-a686-4896-864a-4c3924458709
*
* Covers callbacks registered through POST /team/{team_id}/callback and the Admin UI as well as
* teams still on the deprecated callback_settings shape, resolved from the team's stored metadata
* with the same precedence used at request time. A key-level logging config overrides the team's
* at request time and is not reflected here. Credential-bearing callback_vars are returned masked
* as `***REDACTED***`
*
* Returns {
* "status": "success",
* "data": {
* "team_id": team_id,
* "success_callbacks": team_callback_settings_obj.success_callback,
* "failure_callbacks": team_callback_settings_obj.failure_callback,
* "callback_vars": team_callback_settings_obj.callback_vars,
* },
* }
*/
get: operations["get_team_callbacks_team__team_id__callback_get"];
put?: never;
/**
* Add Team Callbacks
* @description Add a success/failure callback to a team
*
* Use this if if you want different teams to have different success/failure callbacks
*
* Parameters:
* - callback_name (str, required): The name of the callback to add, e.g. "langfuse", "langsmith", "gcs", "newrelic". The value is validated against the callbacks that support team-scoped credentials
* - callback_type (Literal["success", "failure", "success_and_failure"], required): The type of callback to add. One of:
* - "success": Callback for successful LLM calls
* - "failure": Callback for failed LLM calls
* - "success_and_failure": Callback for both successful and failed LLM calls
* - callback_vars (StandardCallbackDynamicParams, required): A dictionary of variables to pass to the callback
* - langfuse_public_key: The public key for the Langfuse callback
* - langfuse_secret_key: The secret key for the Langfuse callback
* - langfuse_secret: The secret for the Langfuse callback
* - langfuse_host: The host for the Langfuse callback
* - langfuse_environment: The tracing environment for the Langfuse callback (lowercase; falls back to LANGFUSE_TRACING_ENVIRONMENT)
* - langfuse_span_scope: For langfuse_otel, "full" (default) sends the whole request trace, "llm_only" sends only the model-call spans
* - capture_message_content: For OTel v2 callbacks, "no_content" or "span_only".
* An explicit value overrides the global setting for this destination; omitted follows the proxy's own callback for that backend, else the global setting.
* "span_only" puts prompt and response content on the destination's spans, "no_content" leaves it out
* - excluded_services: For OTel v2 callbacks, a comma-separated list of datastore names such as "redis,postgres", or "none" to exclude nothing. An explicit value replaces the global exclusions; omitted follows them
* - gcs_bucket_name: The name of the GCS bucket
* - gcs_path_service_account: The path to the GCS service account
* - langsmith_api_key: The API key for the Langsmith callback
* - langsmith_project: The project for the Langsmith callback
* - langsmith_base_url: The base URL for the Langsmith callback
* - newrelic_api_key: The ingest license key for the team's New Relic account; routes both LLM/agent traces and cost metrics to that account. Requires the proxy to run with LITELLM_OTEL_V2=true, otherwise this callback is rejected with a 400
* - newrelic_region: The New Relic region for the team's account ("us" or "eu"), riding the team's own key
*
* Example curl:
* ```
* curl -X POST 'http:/localhost:4000/team/dbe2f686-a686-4896-864a-4c3924458709/callback' -H 'Content-Type: application/json' -H "Authorization: Bearer $LITELLM_MASTER_KEY" -d '{
* "callback_name": "langfuse",
* "callback_type": "success",
* "callback_vars": {"langfuse_public_key": "pk-lf-xxxx1", "langfuse_secret_key": "sk-xxxxx"}
*
* }'
* ```
*
* This means for the team where team_id = dbe2f686-a686-4896-864a-4c3924458709, all LLM calls will be logged to langfuse using the public key pk-lf-xxxx1 and the secret key sk-xxxxx
*/
post: operations["add_team_callbacks_team__team_id__callback_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/team/{team_id}/callback/{callback_name}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
post?: never;
/**
* Delete Team Callback
* @description Remove a single callback from a team
*
* The team's other callbacks stay registered and keep firing. Use this instead of
* POST /team/{team_id}/disable_logging, which clears every callback on the team at once.
*
* Every entry registered under this callback_name is removed, across callback types, so a
* callback registered for both "success" and "failure" is deregistered by one call.
*
* Parameters:
* - team_id (str, required): The unique identifier for the team
* - callback_name (str, required): The name of the callback to remove, matched exactly as it was
* registered with POST /team/{team_id}/callback (e.g. "langfuse", "langsmith", "gcs")
*
* Example curl:
* ```
* curl -X DELETE 'http://localhost:4000/team/dbe2f686-a686-4896-864a-4c3924458709/callback/langsmith' -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*
* Covers callbacks registered through POST /team/{team_id}/callback and the Admin UI. Teams still
* on the deprecated callback_settings metadata shape hold no such entries, so this returns 404 for
* them; POST /team/{team_id}/disable_logging remains the way to clear those.
*
* Returns 404 if the team does not exist, or if callback_name is not registered for the team.
*/
delete: operations["delete_team_callback_team__team_id__callback__callback_name__delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/team/{team_id}/disable_logging": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Disable Team Logging
* @description Disable all logging callbacks for a team
*
* Callbacks registered through POST /team/{team_id}/callback and the Admin UI are cleared, so
* re-enabling logging means registering them again with their callback_vars
*
* Parameters:
* - team_id (str, required): The unique identifier for the team
*
* Example curl:
* ```
* curl -X POST 'http://localhost:4000/team/dbe2f686-a686-4896-864a-4c3924458709/disable_logging' -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*/
post: operations["disable_team_logging_team__team_id__disable_logging_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/team/{team_id}/member/{user_id}/reset_budget": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Reset Team Member Budget Fn
* @description Put a team member back on the team's shared default member budget (`team_member_budget`).
*
* Drops the member's own budget row link so team-wide changes made through /team/update
* reach them again. Leaves the member with no budget when the team has no default. Spend is untouched.
*/
post: operations["reset_team_member_budget_fn_team__team_id__member__user_id__reset_budget_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/team/{team_id}/member/{user_id}/reset_spend": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Reset Team Member Spend Fn
* @description Reset a team member's tracked spend against their per-member budget.
*
* A member's spend is tracked separately from both their own personal
* budget and the team's own budget (LiteLLM_TeamMembership.spend), so
* neither /user/update nor /team/update can clear it: this is the only
* endpoint that does. The cross-pod spend counter and cached membership
* reads are invalidated so the reset takes effect on the member's next
* request rather than waiting on the membership cache's TTL.
*/
post: operations["reset_team_member_spend_fn_team__team_id__member__user_id__reset_spend_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/team/{team_id}/members/me": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Team Member Me
* @description Get the caller's own team-membership row for the given team.
*
* Used by internal users to view their own spend, budget, budget reset
* date, rate limits, and role within a team — without exposing other
* members' data. The caller is resolved from their API key; the path
* `/members/me` always refers to that caller.
*
* Returns 404 if the caller is not a member of the team.
*
* ```
* curl --location 'http://localhost:4000/team/your_team_id/members/me' --header 'Authorization: Bearer your_api_key_here'
* ```
*/
get: operations["team_member_me_team__team_id__members_me_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/telemetry/reports": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Export Telemetry Reports
* @description Stored telemetry reports, oldest first, for installs that keep telemetry local instead of sending it.
* Page with ``after=next_after&after_id=next_after_id`` until ``next_after`` is null. Another worker can commit a
* report just after a newer one is exported, so resume a later export a few windows before the last cursor and skip
* ids already seen
*/
get: operations["export_telemetry_reports_telemetry_reports_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/telemetry/settings": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Telemetry Settings
* @description Which telemetry groups are on, whether env vars control them, where reports go and a last or sample report.
* Nothing is recorded until a destination is chosen by ``LITELLM_TELEMETRY_DESTINATION`` or a saved setting
*/
get: operations["get_telemetry_settings_telemetry_settings_get"];
/**
* Update Telemetry Settings
* @description Store the telemetry groups and destination for every worker, applied from the next report window
*/
put: operations["update_telemetry_settings_telemetry_settings_put"];
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/telemetry/ui_events": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Record Ui Event
* @description Count one Admin UI page view or tab click. Events are dropped unless Page navigation is on
*/
post: operations["record_ui_event_telemetry_ui_events_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/telemetry/ui_events/enabled": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Ui Events Enabled */
get: operations["ui_events_enabled_telemetry_ui_events_enabled_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/test": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Test Endpoint
* @description [DEPRECATED] use `/health/liveliness` instead.
*
* A test endpoint that pings the proxy server to check if it's healthy.
*
* Parameters:
* request (Request): The incoming request.
*
* Returns:
* dict: A dictionary containing the route of the request URL.
*/
get: operations["test_endpoint_test_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/threads": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Create Threads
* @description Create a thread.
*
* API Reference - https://platform.openai.com/docs/api-reference/threads/createThread
*/
post: operations["create_threads_threads_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/threads/{thread_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Thread
* @description Retrieves a thread.
*
* API Reference - https://platform.openai.com/docs/api-reference/threads/getThread
*/
get: operations["get_thread_threads__thread_id__get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/threads/{thread_id}/messages": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Messages
* @description Returns a list of messages for a given thread.
*
* API Reference - https://platform.openai.com/docs/api-reference/messages/listMessages
*/
get: operations["get_messages_threads__thread_id__messages_get"];
put?: never;
/**
* Add Messages
* @description Create a message.
*
* API Reference - https://platform.openai.com/docs/api-reference/messages/createMessage
*/
post: operations["add_messages_threads__thread_id__messages_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/threads/{thread_id}/runs": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Run Thread
* @description Create a run.
*
* API Reference: https://platform.openai.com/docs/api-reference/runs/createRun
*/
post: operations["run_thread_threads__thread_id__runs_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/tinyfish/{endpoint}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Tinyfish Proxy Route
* @description Pass-through for the TinyFish Agent API (goal-based web automation).
*
* Forwarded endpoints:
* - POST /v1/automation/run — run to completion (blocking)
* - POST /v1/automation/run-async — submit a run, poll GET /v1/runs/{id} for the result
* - POST /v1/automation/run-sse — run with SSE progress events
* - GET /v1/runs/{id} — run status / result
* - POST /v1/runs/{id}/cancel — cancel a run
*
* Every other Agent API endpoint (vault, wallet, browser profiles, and the GET /v1/runs
* listing, which would let any caller discover other callers' run ids) returns 403: all
* proxy callers share one upstream key.
*
* Credential lookup order:
* 1. passthrough_endpoint_router (config.yaml deployments with use_in_pass_through)
* 2. TINYFISH_API_KEY environment variable
*
* [Docs](https://docs.litellm.ai/docs/pass_through/tinyfish)
*/
get: operations["tinyfish_proxy_route_tinyfish__endpoint__get"];
put?: never;
/**
* Tinyfish Proxy Route
* @description Pass-through for the TinyFish Agent API (goal-based web automation).
*
* Forwarded endpoints:
* - POST /v1/automation/run — run to completion (blocking)
* - POST /v1/automation/run-async — submit a run, poll GET /v1/runs/{id} for the result
* - POST /v1/automation/run-sse — run with SSE progress events
* - GET /v1/runs/{id} — run status / result
* - POST /v1/runs/{id}/cancel — cancel a run
*
* Every other Agent API endpoint (vault, wallet, browser profiles, and the GET /v1/runs
* listing, which would let any caller discover other callers' run ids) returns 403: all
* proxy callers share one upstream key.
*
* Credential lookup order:
* 1. passthrough_endpoint_router (config.yaml deployments with use_in_pass_through)
* 2. TINYFISH_API_KEY environment variable
*
* [Docs](https://docs.litellm.ai/docs/pass_through/tinyfish)
*/
post: operations["tinyfish_proxy_route_tinyfish__endpoint__post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/token": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Token Endpoint
* @description Accept the authorization code from client and exchange it for OAuth token.
* Supports PKCE flow by forwarding code_verifier to upstream provider.
*
* 1. Call the token endpoint with PKCE parameters
* 2. Store the user's token in the db - and generate a LiteLLM virtual key
* 3. Return the token
* 4. Return a virtual key in this response
*/
post: operations["token_endpoint_token_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/toolset/{toolset_name}/mcp": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Toolset Mcp Route
* @description Namespace a toolset as its own MCP endpoint.
*
* Connecting to /toolset/<name>/mcp exposes exactly the tools defined in
* the toolset. Access is enforced: non-admin API keys must have the toolset
* listed in their object_permission.mcp_toolsets grant list, or the request
* will be rejected with a 403.
*/
get: operations["toolset_mcp_route_toolset__toolset_name__mcp_get"];
/**
* Toolset Mcp Route
* @description Namespace a toolset as its own MCP endpoint.
*
* Connecting to /toolset/<name>/mcp exposes exactly the tools defined in
* the toolset. Access is enforced: non-admin API keys must have the toolset
* listed in their object_permission.mcp_toolsets grant list, or the request
* will be rejected with a 403.
*/
put: operations["toolset_mcp_route_toolset__toolset_name__mcp_put"];
/**
* Toolset Mcp Route
* @description Namespace a toolset as its own MCP endpoint.
*
* Connecting to /toolset/<name>/mcp exposes exactly the tools defined in
* the toolset. Access is enforced: non-admin API keys must have the toolset
* listed in their object_permission.mcp_toolsets grant list, or the request
* will be rejected with a 403.
*/
post: operations["toolset_mcp_route_toolset__toolset_name__mcp_post"];
/**
* Toolset Mcp Route
* @description Namespace a toolset as its own MCP endpoint.
*
* Connecting to /toolset/<name>/mcp exposes exactly the tools defined in
* the toolset. Access is enforced: non-admin API keys must have the toolset
* listed in their object_permission.mcp_toolsets grant list, or the request
* will be rejected with a 403.
*/
delete: operations["toolset_mcp_route_toolset__toolset_name__mcp_delete"];
/**
* Toolset Mcp Route
* @description Namespace a toolset as its own MCP endpoint.
*
* Connecting to /toolset/<name>/mcp exposes exactly the tools defined in
* the toolset. Access is enforced: non-admin API keys must have the toolset
* listed in their object_permission.mcp_toolsets grant list, or the request
* will be rejected with a 403.
*/
options: operations["toolset_mcp_route_toolset__toolset_name__mcp_options"];
/**
* Toolset Mcp Route
* @description Namespace a toolset as its own MCP endpoint.
*
* Connecting to /toolset/<name>/mcp exposes exactly the tools defined in
* the toolset. Access is enforced: non-admin API keys must have the toolset
* listed in their object_permission.mcp_toolsets grant list, or the request
* will be rejected with a 403.
*/
head: operations["toolset_mcp_route_toolset__toolset_name__mcp_head"];
/**
* Toolset Mcp Route
* @description Namespace a toolset as its own MCP endpoint.
*
* Connecting to /toolset/<name>/mcp exposes exactly the tools defined in
* the toolset. Access is enforced: non-admin API keys must have the toolset
* listed in their object_permission.mcp_toolsets grant list, or the request
* will be rejected with a 403.
*/
patch: operations["toolset_mcp_route_toolset__toolset_name__mcp_patch"];
trace?: never;
};
"/transcribe": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Transcribe Sdk Proxy Route
* @description AWS-SDK-shaped pass-through for Amazon Transcribe: point the SDK's `endpoint_url`
* at `/transcribe` and the operation is read from the `X-Amz-Target` header, per the
* AWS JSON 1.1 protocol.
*
* [Docs](https://docs.litellm.ai/docs/pass_through/transcribe)
*/
post: operations["transcribe_sdk_proxy_route_transcribe_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/transcribe/{operation}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Transcribe Proxy Route
* @description Pass-through for the Amazon Transcribe API, e.g. `POST /transcribe/StartTranscriptionJob`.
*
* The request body is forwarded to the AWS JSON 1.1 API and signed with SigV4 using the
* proxy's AWS credentials. Standard jobs are tagged with the calling key's owner so that
* only that owner (or a proxy admin) can read or delete them, and keys other than proxy
* admins may only read media from and write transcripts to the S3 buckets listed in
* `general_settings.transcribe_media_buckets`; account-wide operations
* such as ListTranscriptionJobs are limited to proxy admins. Streaming transcription
* (`transcribestreaming`) uses a separate HTTP/2 event-stream protocol and is not served
* by this route.
*
* [Docs](https://docs.litellm.ai/docs/pass_through/transcribe)
*/
post: operations["transcribe_proxy_route_transcribe__operation__post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/typesafe/{endpoint}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Typesafe Proxy Route
* @description [Docs](https://docs.litellm.ai/docs/pass_through/typesafe)
*/
get: operations["typesafe_proxy_route_typesafe__endpoint__get"];
/**
* Typesafe Proxy Route
* @description [Docs](https://docs.litellm.ai/docs/pass_through/typesafe)
*/
put: operations["typesafe_proxy_route_typesafe__endpoint__put"];
/**
* Typesafe Proxy Route
* @description [Docs](https://docs.litellm.ai/docs/pass_through/typesafe)
*/
post: operations["typesafe_proxy_route_typesafe__endpoint__post"];
/**
* Typesafe Proxy Route
* @description [Docs](https://docs.litellm.ai/docs/pass_through/typesafe)
*/
delete: operations["typesafe_proxy_route_typesafe__endpoint__delete"];
options?: never;
head?: never;
/**
* Typesafe Proxy Route
* @description [Docs](https://docs.litellm.ai/docs/pass_through/typesafe)
*/
patch: operations["typesafe_proxy_route_typesafe__endpoint__patch"];
trace?: never;
};
"/update/default_team_settings": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
/**
* Update Default Team Settings
* @description Update the default team parameters (litellm_settings.default_team_params).
* Applied to every new team for fields not explicitly provided in the create request;
* `models` only applies to teams automatically created via SSO Groups.
*/
patch: operations["update_default_team_settings_update_default_team_settings_patch"];
trace?: never;
};
"/update/internal_user_settings": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
/**
* Update Internal User Settings
* @description Update the default internal user parameters for SSO users.
* These settings will be applied to new users who sign in via SSO.
*/
patch: operations["update_internal_user_settings_update_internal_user_settings_patch"];
trace?: never;
};
"/update/mcp_semantic_filter_settings": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
/**
* Update Mcp Semantic Filter Settings
* @description Update MCP semantic filter settings in database.
* Settings will be picked up by all pods within approximately 10 seconds via background polling.
*/
patch: operations["update_mcp_semantic_filter_settings_update_mcp_semantic_filter_settings_patch"];
trace?: never;
};
"/update/mcp_tool_search_settings": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
/**
* Update Mcp Tool Search Settings
* @description Update `litellm_settings.mcp_tool_search` in the database.
* Settings will be picked up by all pods within approximately 10 seconds via background polling.
*/
patch: operations["update_mcp_tool_search_settings_update_mcp_tool_search_settings_patch"];
trace?: never;
};
"/update/sso_settings": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
/**
* Update Sso Settings
* @description Update SSO configuration by saving to the dedicated SSO table.
*/
patch: operations["update_sso_settings_update_sso_settings_patch"];
trace?: never;
};
"/update/ui_settings": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
/**
* Update Ui Settings
* @description Update UI-specific configuration flags.
* Only proxy admins are allowed to modify these settings.
*/
patch: operations["update_ui_settings_update_ui_settings_patch"];
trace?: never;
};
"/update/ui_theme_settings": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
/**
* Update Ui Theme Settings
* @description Update UI theme configuration.
* Updates logo settings for the admin UI.
*/
patch: operations["update_ui_theme_settings_update_ui_theme_settings_patch"];
trace?: never;
};
"/update/user_banner": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
/**
* Update User Banner
* @description Publish, edit, or unpublish the dashboard banner.
* Only proxy admins are allowed to modify it.
*/
patch: operations["update_user_banner_update_user_banner_patch"];
trace?: never;
};
"/update/websearch_interception_settings": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
/**
* Update Websearch Interception Settings
* @description Update web search interception settings in database.
*
* Settings will be picked up by all pods within approximately 10 seconds via background polling.
*/
patch: operations["update_websearch_interception_settings_update_websearch_interception_settings_patch"];
trace?: never;
};
"/upload/logo": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Upload Logo
* @description Upload a custom logo for the admin UI.
* Accepts image files (PNG, JPG, JPEG, SVG) and stores them for use in the UI.
*/
post: operations["upload_logo_upload_logo_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/usage/ai/chat": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Usage Ai Chat
* @description AI chat about usage data. Streams SSE events with the AI response.
* The AI agent has access to tools that query aggregated daily activity data.
*/
post: operations["usage_ai_chat_usage_ai_chat_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/user/available_roles": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Ui Get Available Role
* @description Endpoint used by Admin UI to show all available roles to assign a user
* return {
* "proxy_admin": {
* "description": "Proxy Admin role",
* "ui_label": "Admin"
* }
* }
*/
get: operations["ui_get_available_role_user_available_roles_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/user/available_users": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Available Enterprise Users
* @description For keys with `max_users` set, return the list of users that are allowed to use the key.
*/
get: operations["available_enterprise_users_user_available_users_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/user/bulk_update": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Bulk User Update
* @description Bulk update multiple users at once.
*
* This endpoint allows updating multiple users in a single request. Each user update
* is processed independently - if some updates fail, others will still succeed.
*
* Parameters:
* - users: Optional[List[UpdateUserRequest]] - List of specific user update requests
* - all_users: Optional[bool] - Set to true to update all users in the system
* - user_updates: Optional[UpdateUserRequest] - Updates to apply when all_users=True
*
* Returns:
* - results: List of individual update results
* - total_requested: Total number of users requested for update
* - successful_updates: Number of successful updates
* - failed_updates: Number of failed updates
*
* Example request for specific users:
* ```bash
* curl --location 'http://0.0.0.0:4000/user/bulk_update' --header "Authorization: Bearer $LITELLM_MASTER_KEY" --header 'Content-Type: application/json' --data '{
* "users": [
* {
* "user_id": "user1",
* "user_role": "internal_user",
* "max_budget": 100.0
* },
* {
* "user_email": "user2@example.com",
* "user_role": "internal_user_viewer",
* "max_budget": 50.0
* }
* ]
* }'
* ```
*
* Example request for all users:
* ```bash
* curl --location 'http://0.0.0.0:4000/user/bulk_update' --header "Authorization: Bearer $LITELLM_MASTER_KEY" --header 'Content-Type: application/json' --data '{
* "all_users": true,
* "user_updates": {
* "user_role": "internal_user",
* "max_budget": 50.0
* }
* }'
* ```
*/
post: operations["bulk_user_update_user_bulk_update_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/user/daily/activity": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get User Daily Activity
* @description [BETA] This is a beta endpoint. It will change.
*
* Meant to optimize querying spend data for analytics for a user.
*
* Reads daily spend records that only ever accumulate and are never affected by budget
* resets. Their total can legitimately exceed the `spend` field returned by
* `/v2/user/info`, which is a running budget counter that every budget reset sets back
* to zero (or to the overage above `max_budget` when `budget_rollover` is enabled).
*
* Returns:
* (by date)
* - spend
* - prompt_tokens
* - completion_tokens
* - cache_read_input_tokens
* - cache_creation_input_tokens
* - total_tokens
* - api_requests
* - breakdown by model, api_key, provider
*/
get: operations["get_user_daily_activity_user_daily_activity_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/user/daily/activity/aggregated": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get User Daily Activity Aggregated
* @description Aggregated analytics for a user's daily activity without pagination.
* Returns the same response shape as the paginated endpoint with page metadata set to single-page.
*
* Reads daily spend records that only ever accumulate and are never affected by budget
* resets. Their total can legitimately exceed the `spend` field returned by
* `/v2/user/info`, which is a running budget counter that every budget reset sets back
* to zero (or to the overage above `max_budget` when `budget_rollover` is enabled).
*/
get: operations["get_user_daily_activity_aggregated_user_daily_activity_aggregated_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/user/daily/activity/aggregated/cache_leakage_keys": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get User Daily Activity Cache Leakage Keys */
get: operations["get_user_daily_activity_cache_leakage_keys_user_daily_activity_aggregated_cache_leakage_keys_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/user/daily/activity/aggregated/keys": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get User Daily Activity Aggregated Keys */
get: operations["get_user_daily_activity_aggregated_keys_user_daily_activity_aggregated_keys_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/user/daily/activity/aggregated/model_top_keys": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get User Daily Activity Model Top Keys */
get: operations["get_user_daily_activity_model_top_keys_user_daily_activity_aggregated_model_top_keys_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/user/daily/activity/aggregated/search": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get User Daily Activity Aggregated Search */
get: operations["get_user_daily_activity_aggregated_search_user_daily_activity_aggregated_search_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/user/daily/activity/aggregated/users": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get User Daily Activity Aggregated Users */
get: operations["get_user_daily_activity_aggregated_users_user_daily_activity_aggregated_users_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/user/daily/activity/export": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get User Daily Activity Export */
get: operations["get_user_daily_activity_export_user_daily_activity_export_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/user/delete": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Delete User
* @description delete user and associated user keys
*
* ```
* curl --location 'http://0.0.0.0:4000/user/delete'
* --header "Authorization: Bearer $LITELLM_MASTER_KEY"
* --header 'Content-Type: application/json'
* --data-raw '{
* "user_ids": ["45e3e396-ee08-4a61-a88e-16b3ce7e0849"]
* }'
* ```
*
* Parameters:
* - user_ids: List[str] - The list of user id's to be deleted.
*/
post: operations["delete_user_user_delete_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/user/filter/ui": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Ui View Users
* @description Filter users based on partial match of user_id or email, or combined ``search``, with pagination.
*
* Behaviour depends on the ``scope_user_search_to_org`` UI-setting flag
* (stored in the ``litellm_uisettings`` table):
*
* * **Flag OFF (default):** any authenticated user can search all users.
* * **Flag ON:**
* - Proxy admins see all users.
* - Org admins see only users in their org(s).
* - Team admins for an org-bound team see users in that org.
* - Others receive a 403.
*/
get: operations["ui_view_users_user_filter_ui_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/user/info": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* User Info
* @description [10/07/2024]
* Note: To get all users (+pagination), use `/user/list` endpoint.
*
*
* Use this to get user information. (user row + all user key info)
*
* Example request
* ```
* curl -X GET 'http://localhost:4000/user/info?user_id=krrish7%40berri.ai' --header "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*/
get: operations["user_info_user_info_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/user/list": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Users
* @description Get a paginated list of users with filtering and sorting options.
*
* Parameters:
* role: Optional[str]
* Filter users by role. Can be one of:
* - proxy_admin
* - proxy_admin_viewer
* - internal_user
* - internal_user_viewer
* user_ids: Optional[str]
* Get list of users by user_ids. Comma separated list of user_ids.
* sso_ids: Optional[str]
* Get list of users by sso_ids. Comma separated list of sso_ids.
* user_email: Optional[str]
* Filter users by partial email match
* search: Optional[str]
* Combined search: matches users whose user_id or user_email contains the value (case-insensitive)
* team: Optional[str]
* Filter users by team id. Will match if user has this team in their teams array.
* page: int
* The page number to return
* page_size: int
* The number of items per page
* sort_by: Optional[str]
* Column to sort by (e.g. 'user_id', 'user_email', 'created_at', 'spend')
* sort_order: Optional[str]
* Sort order ('asc' or 'desc')
*/
get: operations["get_users_user_list_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/user/new": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* New User
* @description Use this to create a new INTERNAL user with a budget.
* Internal Users can access LiteLLM Admin UI to make keys, request access to models.
* This creates a new user and generates a new api key for the new user. The new api key is returned.
*
* Returns user id, budget + new key.
*
* Parameters:
* - user_id: Optional[str] - Specify a user id. If not set, a unique id will be generated.
* - user_alias: Optional[str] - A descriptive name for you to know who this user id refers to.
* - teams: Optional[list] - specify a list of team id's a user belongs to.
* - user_email: Optional[str] - Specify a user email.
* - send_invite_email: Optional[bool] - Specify if an invite email should be sent.
* - user_role: Optional[str] - Specify a user role - "proxy_admin", "proxy_admin_viewer", "internal_user", "internal_user_viewer", "team", "customer". Info about each role here: `https://github.com/BerriAI/litellm/litellm/proxy/_types.py#L20`
* - max_budget: Optional[float] - Specify max budget for a given user.
* - budget_duration: Optional[str] - Budget is reset at the end of specified duration. If not set, budget is never reset. You can set duration as seconds ("30s"), minutes ("30m"), hours ("30h"), days ("30d"), months ("1mo").
* - models: Optional[list] - Model_name's a user is allowed to call. (if empty, key is allowed to call all models). Set to ['no-default-models'] to block all model access. Restricting user to only team-based model access.
* - tpm_limit: Optional[int] - Specify tpm limit for a given user (Tokens per minute)
* - rpm_limit: Optional[int] - Specify rpm limit for a given user (Requests per minute)
* - auto_create_key: bool - Default=True. Flag used for returning a key as part of the /user/new response
* - aliases: Optional[dict] - Model aliases for the user - [Docs](https://litellm.vercel.app/docs/proxy/virtual_keys#model-aliases)
* - config: Optional[dict] - [DEPRECATED PARAM] User-specific config.
* - allowed_cache_controls: Optional[list] - List of allowed cache control values. Example - ["no-cache", "no-store"]. See all values - https://docs.litellm.ai/docs/proxy/caching#turn-on--off-caching-per-request-
* - blocked: Optional[bool] - [Not Implemented Yet] Whether the user is blocked.
* - guardrails: Optional[List[str]] - [Not Implemented Yet] List of active guardrails for the user
* - policies: Optional[List[str]] - List of policy names to apply to the user. Policies define guardrails, conditions, and inheritance rules.
* - permissions: Optional[dict] - [Not Implemented Yet] User-specific permissions, eg. turning off pii masking.
* - metadata: Optional[dict] - Metadata for user, store information for user. Example metadata = {"team": "core-infra", "app": "app2", "email": "[email protected]" }
* - max_parallel_requests: Optional[int] - Rate limit a user based on the number of parallel requests. Raises 429 error, if user's parallel requests > x.
* - model_max_budget: Optional[dict] - Model-specific max budget for user. [Docs](https://docs.litellm.ai/docs/proxy/users#add-model-specific-budgets-to-keys)
* - budget_fallbacks: Optional[Dict[str, List[str]]] - Per-model fallback chain tried in order when that model's own `model_max_budget` is exceeded, e.g. {"gpt-4o": ["gpt-4o-mini"]}.
* - model_rpm_limit: Optional[float] - Model-specific rpm limit for user. [Docs](https://docs.litellm.ai/docs/proxy/users#add-model-specific-limits-to-keys)
* - mcp_rpm_limit: Optional[dict] - Per-MCP-server rpm limit, keyed by MCP server name {"github": 100, "slack": 200}. Enforced for keys and teams only; values set on a user are stored but not enforced per user.
* - tag_rpm_limit: Optional[dict] - Per-request-tag rpm limit, keyed by request tag {"cell-1": 1000, "cell-2": 500}. Enforced for keys only; values set on a user are stored but not enforced per user.
* - model_tpm_limit: Optional[float] - Model-specific tpm limit for user. [Docs](https://docs.litellm.ai/docs/proxy/users#add-model-specific-limits-to-keys)
* - spend: Optional[float] - Amount spent by user. Default is 0. Will be updated by proxy whenever user is used. You can set duration as seconds ("30s"), minutes ("30m"), hours ("30h"), days ("30d"), months ("1mo").
* - agent_id: Optional[str] - The agent id associated with the user.
* - team_id: Optional[str] - [DEPRECATED PARAM] The team id of the user. Default is None.
* - duration: Optional[str] - Duration for the key auto-created on `/user/new`. Default is None.
* - key_alias: Optional[str] - Alias for the key auto-created on `/user/new`. Default is None.
* - sso_user_id: Optional[str] - The id of the user in the SSO provider.
* - object_permission: Optional[LiteLLM_ObjectPermissionBase] - internal user-specific object permission. Example - {"vector_stores": ["vector_store_1"], "mcp_servers": ["github"], "mcp_tool_permissions": {"github": ["list_issues"]}}. The MCP grants act as a ceiling on every key this user holds. IF null or {} then no object permission.
* - prompts: Optional[List[str]] - List of allowed prompts for the user. If specified, the user will only be able to use these specific prompts.
* - organizations: List[str] - List of organization id's the user is a member of
* - budget_limits: Optional[list] - List of concurrent budget windows for the user. Each window specifies a budget_limit, time_period, and optional budget_duration. Example - [{"budget_limit": 10.0, "time_period": "1d"}, {"budget_limit": 50.0, "time_period": "7d"}].
* - password: Optional[str] - Not supported; any value is rejected with a 422. Users set their own password through an invitation link (POST /invitation/new).
* Returns:
* - key: (str) The generated api key for the user
* - expires: (datetime) Datetime object for when key expires.
* - user_id: (str) Unique user id - used for tracking spend across multiple keys for same user id.
* - max_budget: (float|None) Max budget for given user.
*
* Usage Example
*
* ```shell
* curl -X POST "http://localhost:4000/user/new" -H "Content-Type: application/json" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -d '{
* "username": "new_user",
* "email": "new_user@example.com"
* }'
* ```
*/
post: operations["new_user_user_new_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/user/password/change": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Change Password
* @description Change the calling user's own password.
*
* Only callable with the dashboard session issued by a username/password
* login; SSO sessions and virtual keys are rejected with 403. Requires the
* current password. The new password must differ from the
* current one and satisfy the configured password policy
* (`general_settings.password_policy_*`: minimum length, character classes,
* and, when enabled, breached-password screening via haveibeenpwned.com).
* A successful change lifts any pending forced password reset
* (`password_reset_required`) on the account.
*
* Parameters:
* - current_password: str - The user's current password.
* - new_password: str - The password to change to.
*/
post: operations["change_password_user_password_change_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/user/spend/report": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get User Spend Report
* @description Get spend for the calling user over a date range, grouped by api_key with a per-model breakdown.
*
* Same row shape as `/global/spend/report?internal_user_id=...`, but callable by
* any key with a user: non-admin callers are always scoped to their own user_id,
* while proxy admins may pass `?internal_user_id=` to view any user.
*/
get: operations["get_user_spend_report_user_spend_report_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/user/update": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* User Update
* @description Example curl
*
* ```
* curl --location 'http://0.0.0.0:4000/user/update' --header "Authorization: Bearer $LITELLM_MASTER_KEY" --header 'Content-Type: application/json' --data '{
* "user_id": "test-litellm-user-4",
* "user_role": "proxy_admin_viewer"
* }'
* ```
*
* Parameters:
* - user_id: Optional[str] - Specify a user id. If not set, a unique id will be generated.
* - user_email: Optional[str] - Specify a user email.
* - password: Optional[str] - Set the user's password (admin only). Must satisfy the configured password policy. The user is required to change it at their next login. Users change their own password with POST /user/password/change.
* - user_alias: Optional[str] - A descriptive name for you to know who this user id refers to.
* - teams: Optional[list] - specify a list of team id's a user belongs to.
* - send_invite_email: Optional[bool] - Specify if an invite email should be sent.
* - user_role: Optional[str] - Specify a user role - "proxy_admin", "proxy_admin_viewer", "internal_user", "internal_user_viewer", "team", "customer". Info about each role here: `https://github.com/BerriAI/litellm/litellm/proxy/_types.py#L20`
* - max_budget: Optional[float] - Specify max budget for a given user.
* - budget_duration: Optional[str] - Budget is reset at the end of specified duration. If not set, budget is never reset. You can set duration as seconds ("30s"), minutes ("30m"), hours ("30h"), days ("30d"), months ("1mo").
* - models: Optional[list] - Model_name's a user is allowed to call. (if empty, key is allowed to call all models)
* - tpm_limit: Optional[int] - Specify tpm limit for a given user (Tokens per minute)
* - rpm_limit: Optional[int] - Specify rpm limit for a given user (Requests per minute)
* - auto_create_key: bool - Default=True. Flag used for returning a key as part of the /user/new response
* - aliases: Optional[dict] - Model aliases for the user - [Docs](https://litellm.vercel.app/docs/proxy/virtual_keys#model-aliases)
* - config: Optional[dict] - [DEPRECATED PARAM] User-specific config.
* - allowed_cache_controls: Optional[list] - List of allowed cache control values. Example - ["no-cache", "no-store"]. See all values - https://docs.litellm.ai/docs/proxy/caching#turn-on--off-caching-per-request-
* - blocked: Optional[bool] - [Not Implemented Yet] Whether the user is blocked.
* - guardrails: Optional[List[str]] - [Not Implemented Yet] List of active guardrails for the user
* - policies: Optional[List[str]] - List of policy names to apply to the user. Policies define guardrails, conditions, and inheritance rules.
* - permissions: Optional[dict] - [Not Implemented Yet] User-specific permissions, eg. turning off pii masking.
* - metadata: Optional[dict] - Metadata for user, store information for user. Example metadata = {"team": "core-infra", "app": "app2", "email": "[email protected]" }
* - max_parallel_requests: Optional[int] - Rate limit a user based on the number of parallel requests. Raises 429 error, if user's parallel requests > x.
* - model_max_budget: Optional[dict] - Model-specific max budget for user. [Docs](https://docs.litellm.ai/docs/proxy/users#add-model-specific-budgets-to-keys)
* - budget_fallbacks: Optional[Dict[str, List[str]]] - Per-model fallback chain tried in order when that model's own `model_max_budget` is exceeded, e.g. {"gpt-4o": ["gpt-4o-mini"]}.
* - model_rpm_limit: Optional[float] - Model-specific rpm limit for user. [Docs](https://docs.litellm.ai/docs/proxy/users#add-model-specific-limits-to-keys)
* - mcp_rpm_limit: Optional[dict] - Per-MCP-server rpm limit, keyed by MCP server name {"github": 100, "slack": 200}. Enforced for keys and teams only; values set on a user are stored but not enforced per user.
* - tag_rpm_limit: Optional[dict] - Per-request-tag rpm limit, keyed by request tag {"cell-1": 1000, "cell-2": 500}. Enforced for keys only; values set on a user are stored but not enforced per user.
* - model_tpm_limit: Optional[float] - Model-specific tpm limit for user. [Docs](https://docs.litellm.ai/docs/proxy/users#add-model-specific-limits-to-keys)
* - spend: Optional[float] - Amount spent by user. Default is 0. Will be updated by proxy whenever user is used. You can set duration as seconds ("30s"), minutes ("30m"), hours ("30h"), days ("30d"), months ("1mo").
* - agent_id: Optional[str] - The agent id associated with the user.
* - team_id: Optional[str] - [DEPRECATED PARAM] The team id of the user. Default is None.
* - duration: Optional[str] - [NOT IMPLEMENTED].
* - key_alias: Optional[str] - [NOT IMPLEMENTED].
* - object_permission: Optional[LiteLLM_ObjectPermissionBase] - internal user-specific object permission. Example - {"vector_stores": ["vector_store_1"], "mcp_servers": ["github"], "mcp_tool_permissions": {"github": ["list_issues"]}}. The MCP grants act as a ceiling on every key this user holds. IF null or {} then no object permission.
* - prompts: Optional[List[str]] - List of allowed prompts for the user. If specified, the user will only be able to use these specific prompts.
* - budget_limits: Optional[list] - List of concurrent budget windows for the user. Each window specifies a budget_limit, time_period, and optional budget_duration. Example - [{"budget_limit": 10.0, "time_period": "1d"}, {"budget_limit": 50.0, "time_period": "7d"}].
*/
post: operations["user_update_user_update_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/utils/dotprompt_json_converter": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Convert Prompt File To Json
* @description Convert a .prompt file to JSON format.
*
* This endpoint accepts a .prompt file upload and returns the equivalent JSON representation
* that can be stored in a database or used programmatically.
*
* Returns the JSON structure with 'content' and 'metadata' fields.
*/
post: operations["convert_prompt_file_to_json_utils_dotprompt_json_converter_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/utils/model_info": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Model Info Lookup
* @description Returns the model cost map entry (token limits, pricing, supports_* capabilities) for any model
* in the cost map, whether or not it is registered on this proxy. `model_info` carries every
* field of the raw cost map entry plus the typed fields `litellm.get_model_info` derives from it
* (`key`, `supported_openai_params`).
*
* Example curl:
* ```
* curl -X GET --location 'http://localhost:4000/utils/model_info?model=gpt-4o&custom_llm_provider=openai' --header "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*/
get: operations["model_info_lookup_utils_model_info_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/utils/supported_openai_params": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Supported Openai Params
* @description Returns supported openai params for a given litellm model name
*
* e.g. `gpt-4` vs `gpt-3.5-turbo`
*
* Example curl:
* ```
* curl -X GET --location 'http://localhost:4000/utils/supported_openai_params?model=gpt-3.5-turbo-16k' --header "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*/
get: operations["supported_openai_params_utils_supported_openai_params_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/utils/test_policies_and_guardrails": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Test Policies And Guardrails
* @description Apply policies and/or guardrails to inputs (for compliance UI testing).
*
* Use inputs_list for batch testing: each input is processed as a separate call so
* per-input block/allow and errors are returned.
*
* Use inputs for a single call (legacy).
*/
post: operations["test_policies_and_guardrails_utils_test_policies_and_guardrails_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/utils/token_counter": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Token Counter
* @description Args:
* request: TokenCountRequest
* call_endpoint: bool - When set to "True" it will call the token counting endpoint - e.g Anthropic or Google AI Studio Token Counting APIs.
*
* Returns:
* TokenCountResponse
*/
post: operations["token_counter_utils_token_counter_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/utils/transform_request": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/** Transform Request */
post: operations["transform_request_utils_transform_request_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/a2a/discover": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Discover Agent Card
* @description Fetch the upstream agent's well-known card so the UI can show the admin
* which skills/capabilities the agent exposes.
*
* Only proxy admins can call this — the UI uses it during agent registration,
* and we don't want arbitrary keys probing internal URLs.
*
* Example:
* ```bash
* curl -X POST "http://localhost:4000/v1/a2a/discover" \
* -H "Authorization: Bearer <admin_key>" \
* -H "Content-Type: application/json" \
* -d '{"url": "https://upstream-agent.example.com"}'
* ```
*/
post: operations["discover_agent_card_v1_a2a_discover_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/a2a/{agent_id}/message/send": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Invoke Agent A2A
* @description Invoke an agent using the A2A protocol (JSON-RPC 2.0).
*
* Supported methods:
* - message/send: Send a message and get a response
* - message/stream: Send a message and stream the response
*/
post: operations["invoke_agent_a2a_v1_a2a__agent_id__message_send_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/access_group": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** List Access Groups */
get: operations["list_access_groups_v1_access_group_get"];
put?: never;
/** Create Access Group */
post: operations["create_access_group_v1_access_group_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/access_group/{access_group_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get Access Group */
get: operations["get_access_group_v1_access_group__access_group_id__get"];
/** Update Access Group */
put: operations["update_access_group_v1_access_group__access_group_id__put"];
post?: never;
/** Delete Access Group */
delete: operations["delete_access_group_v1_access_group__access_group_id__delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/agents": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Agents
* @description Example usage:
* ```
* curl -X GET "http://localhost:4000/v1/agents" -H "Content-Type: application/json" -H "Authorization: Bearer your-key" ```
*
* Pass `?health_check=true` to filter out agents whose URL is unreachable:
* ```
* curl -X GET "http://localhost:4000/v1/agents?health_check=true" -H "Content-Type: application/json" -H "Authorization: Bearer your-key" ```
*
* Pass `?query=<task>` to get the best matching agents ranked by semantic similarity:
* ```
* curl -X GET "http://localhost:4000/v1/agents?query=translate+a+PDF+document&top_k=5" -H "Content-Type: application/json" -H "Authorization: Bearer your-key" ```
*
* Returns: List[AgentResponse]
*/
get: operations["get_agents_v1_agents_get"];
put?: never;
/**
* Create Agent
* @description Create a new agent
*
* Example Request:
* ```bash
* curl -X POST "http://localhost:4000/v1/agents" \
* -H "Authorization: Bearer <your_api_key>" \
* -H "Content-Type: application/json" \
* -d '{
* "agent_name": "my-custom-agent",
* "agent_card_params": {
* "protocolVersion": "1.0",
* "name": "Hello World Agent",
* "description": "Just a hello world agent",
* "url": "http://localhost:9999/",
* "version": "1.0.0",
* "defaultInputModes": ["text"],
* "defaultOutputModes": ["text"],
* "capabilities": {
* "streaming": true
* },
* "skills": [
* {
* "id": "hello_world",
* "name": "Returns hello world",
* "description": "just returns hello world",
* "tags": ["hello world"],
* "examples": ["hi", "hello world"]
* }
* ]
* },
* "litellm_params": {
* "make_public": true
* }
* }'
* ```
*/
post: operations["create_agent_v1_agents_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/agents/identity/providers": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get Agent Identity Providers */
get: operations["get_agent_identity_providers_v1_agents_identity_providers_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/agents/make_public": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Make Agents Public
* @description Make multiple agents publicly discoverable
*
* Example Request:
* ```bash
* curl -X POST "http://localhost:4000/v1/agents/make_public" \
* -H "Authorization: Bearer <your_api_key>" \
* -H "Content-Type: application/json" \
* -d '{
* "agent_ids": ["123e4567-e89b-12d3-a456-426614174000", "123e4567-e89b-12d3-a456-426614174001"]
* }'
* ```
*
* Example Response:
* ```json
* {
* "agent_id": "123e4567-e89b-12d3-a456-426614174000",
* "agent_name": "my-custom-agent",
* "litellm_params": {
* "make_public": true
* },
* "agent_card_params": {...},
* "created_at": "2025-11-15T10:30:00Z",
* "updated_at": "2025-11-15T10:35:00Z",
* "created_by": "user123",
* "updated_by": "user123"
* }
* ```
*/
post: operations["make_agents_public_v1_agents_make_public_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/agents/{agent_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Agent By Id
* @description Get a specific agent by ID
*
* Example Request:
* ```bash
* curl -X GET "http://localhost:4000/v1/agents/123e4567-e89b-12d3-a456-426614174000" \
* -H "Authorization: Bearer <your_api_key>"
* ```
*/
get: operations["get_agent_by_id_v1_agents__agent_id__get"];
/**
* Update Agent
* @description Update an existing agent
*
* Example Request:
* ```bash
* curl -X PUT "http://localhost:4000/v1/agents/123e4567-e89b-12d3-a456-426614174000" \
* -H "Authorization: Bearer <your_api_key>" \
* -H "Content-Type: application/json" \
* -d '{
* "agent_name": "updated-agent",
* "agent_card_params": {
* "protocolVersion": "1.0",
* "name": "Updated Agent",
* "description": "Updated description",
* "url": "http://localhost:9999/",
* "version": "1.1.0",
* "defaultInputModes": ["text"],
* "defaultOutputModes": ["text"],
* "capabilities": {
* "streaming": true
* },
* "skills": []
* },
* "litellm_params": {
* "make_public": false
* }
* }'
* ```
*/
put: operations["update_agent_v1_agents__agent_id__put"];
post?: never;
/**
* Delete Agent
* @description Delete an agent
*
* Example Request:
* ```bash
* curl -X DELETE "http://localhost:4000/v1/agents/123e4567-e89b-12d3-a456-426614174000" \
* -H "Authorization: Bearer <your_api_key>"
* ```
*
* Example Response:
* ```json
* {
* "message": "Agent 123e4567-e89b-12d3-a456-426614174000 deleted successfully"
* }
* ```
*/
delete: operations["delete_agent_v1_agents__agent_id__delete"];
options?: never;
head?: never;
/**
* Patch Agent
* @description Update an existing agent
*
* Example Request:
* ```bash
* curl -X PATCH "http://localhost:4000/v1/agents/123e4567-e89b-12d3-a456-426614174000" \
* -H "Authorization: Bearer <your_api_key>" \
* -H "Content-Type: application/json" \
* -d '{
* "agent_name": "updated-agent",
* "agent_card_params": {
* "protocolVersion": "1.0",
* "name": "Updated Agent",
* "description": "Updated description",
* "url": "http://localhost:9999/",
* "version": "1.1.0",
* "defaultInputModes": ["text"],
* "defaultOutputModes": ["text"],
* "capabilities": {
* "streaming": true
* },
* "skills": []
* },
* "litellm_params": {
* "make_public": false
* }
* }'
* ```
*/
patch: operations["patch_agent_v1_agents__agent_id__patch"];
trace?: never;
};
"/v1/agents/{agent_id}/identity": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get Agent Identity Status */
get: operations["get_agent_identity_status_v1_agents__agent_id__identity_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/agents/{agent_id}/kill_switch": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Trigger Agent Kill Switch
* @description Fire the agent's configured kill switch webhook. Proxy admin only.
*
* LiteLLM only makes the configured HTTP call and reports what came back; it
* does not change the agent's state in LiteLLM. Returns 200 when the webhook
* answered 2xx, 502 with the same result body otherwise. Every attempt is
* written to the audit log as a `kill_switch_fired` row against the agent.
*
* Example Request:
* ```bash
* curl -X POST "http://localhost:4000/v1/agents/123e4567-e89b-12d3-a456-426614174000/kill_switch" \
* -H "Authorization: Bearer <your_api_key>"
* ```
*/
post: operations["trigger_agent_kill_switch_v1_agents__agent_id__kill_switch_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/agents/{agent_id}/make_public": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Make Agent Public
* @description Make an agent publicly discoverable
*
* Example Request:
* ```bash
* curl -X POST "http://localhost:4000/v1/agents/123e4567-e89b-12d3-a456-426614174000/make_public" \
* -H "Authorization: Bearer <your_api_key>" \
* -H "Content-Type: application/json"
* ```
*
* Example Response:
* ```json
* {
* "agent_id": "123e4567-e89b-12d3-a456-426614174000",
* "agent_name": "my-custom-agent",
* "litellm_params": {
* "make_public": true
* },
* "agent_card_params": {...},
* "created_at": "2025-11-15T10:30:00Z",
* "updated_at": "2025-11-15T10:35:00Z",
* "created_by": "user123",
* "updated_by": "user123"
* }
* ```
*/
post: operations["make_agent_public_v1_agents__agent_id__make_public_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/assistants": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Assistants
* @description Returns a list of assistants.
*
* API Reference docs - https://platform.openai.com/docs/api-reference/assistants/listAssistants
*/
get: operations["get_assistants_v1_assistants_get"];
put?: never;
/**
* Create Assistant
* @description Create assistant
*
* API Reference docs - https://platform.openai.com/docs/api-reference/assistants/createAssistant
*/
post: operations["create_assistant_v1_assistants_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/assistants/{assistant_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
post?: never;
/**
* Delete Assistant
* @description Delete assistant
*
* API Reference docs - https://platform.openai.com/docs/api-reference/assistants/createAssistant
*/
delete: operations["delete_assistant_v1_assistants__assistant_id__delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/audio/speech": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Audio Speech
* @description Same params as:
*
* https://platform.openai.com/docs/api-reference/audio/createSpeech
*/
post: operations["audio_speech_v1_audio_speech_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/audio/transcriptions": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Audio Transcriptions
* @description Same params as:
*
* https://platform.openai.com/docs/api-reference/audio/createTranscription?lang=curl
*/
post: operations["audio_transcriptions_v1_audio_transcriptions_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/batches": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List Batches
* @description Lists
* This is the equivalent of GET https://api.openai.com/v1/batches/
* Supports Identical Params as: https://platform.openai.com/docs/api-reference/batch/list
*
* Example Curl
* ```
* curl http://localhost:4000/v1/batches?limit=2 -H "Authorization: Bearer $LITELLM_MASTER_KEY" -H "Content-Type: application/json"
* ```
*/
get: operations["list_batches_v1_batches_get"];
put?: never;
/**
* Create Batch
* @description Create large batches of API requests for asynchronous processing.
* This is the equivalent of POST https://api.openai.com/v1/batch
* Supports Identical Params as: https://platform.openai.com/docs/api-reference/batch
*
* Example Curl
* ```
* curl http://localhost:4000/v1/batches -H "Authorization: Bearer $LITELLM_MASTER_KEY" -H "Content-Type: application/json" -d '{
* "input_file_id": "file-abc123",
* "endpoint": "/v1/chat/completions",
* "completion_window": "24h"
* }'
* ```
*/
post: operations["create_batch_v1_batches_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/batches/{batch_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Retrieve Batch
* @description Retrieves a batch.
* This is the equivalent of GET https://api.openai.com/v1/batches/{batch_id}
* Supports Identical Params as: https://platform.openai.com/docs/api-reference/batch/retrieve
*
* Example Curl
* ```
* curl http://localhost:4000/v1/batches/batch_abc123 -H "Authorization: Bearer $LITELLM_MASTER_KEY" -H "Content-Type: application/json"
* ```
*/
get: operations["retrieve_batch_v1_batches__batch_id__get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/batches/{batch_id}/cancel": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Cancel Batch
* @description Cancel a batch.
* This is the equivalent of POST https://api.openai.com/v1/batches/{batch_id}/cancel
*
* Supports Identical Params as: https://platform.openai.com/docs/api-reference/batch/cancel
*
* Example Curl
* ```
* curl http://localhost:4000/v1/batches/batch_abc123/cancel -H "Authorization: Bearer $LITELLM_MASTER_KEY" -H "Content-Type: application/json" -X POST
*
* ```
*/
post: operations["cancel_batch_v1_batches__batch_id__cancel_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/chat/completions": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Chat Completion
* @description Follows the exact same API spec as `OpenAI's Chat API https://platform.openai.com/docs/api-reference/chat`
*
* ```bash
* curl -X POST http://localhost:4000/v1/chat/completions
* -H "Content-Type: application/json"
* -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* -d '{
* "model": "gpt-4o",
* "messages": [
* {
* "role": "user",
* "content": "Hello!"
* }
* ]
* }'
* ```
*/
post: operations["chat_completion_v1_chat_completions_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/completions": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Completion
* @description Follows the exact same API spec as `OpenAI's Completions API https://platform.openai.com/docs/api-reference/completions`
*
* ```bash
* curl -X POST http://localhost:4000/v1/completions
* -H "Content-Type: application/json"
* -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* -d '{
* "model": "gpt-3.5-turbo-instruct",
* "prompt": "Once upon a time",
* "max_tokens": 50,
* "temperature": 0.7
* }'
* ```
*/
post: operations["completion_v1_completions_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/containers": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List Containers
* @description Container list endpoint for retrieving a list of containers.
*
* Follows the OpenAI Containers API spec:
* https://platform.openai.com/docs/api-reference/containers
*
* Example:
* ```bash
* curl -X GET "http://localhost:4000/v1/containers?limit=20&order=desc" -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*
* Or specify provider via header or query param:
* ```bash
* curl -X GET "http://localhost:4000/v1/containers?custom_llm_provider=azure" -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*/
get: operations["list_containers_v1_containers_get"];
put?: never;
/**
* Create Container
* @description Container creation endpoint for creating new containers.
*
* Follows the OpenAI Containers API spec:
* https://platform.openai.com/docs/api-reference/containers
*
* Example:
* ```bash
* curl -X POST "http://localhost:4000/v1/containers" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -H "Content-Type: application/json" -d '{
* "name": "My Container",
* "expires_after": {
* "anchor": "last_active_at",
* "minutes": 20
* }
* }'
* ```
*
* Or specify provider via header:
* ```bash
* curl -X POST "http://localhost:4000/v1/containers" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -H "custom-llm-provider: azure" -H "Content-Type: application/json" -d '{
* "name": "My Container"
* }'
* ```
*/
post: operations["create_container_v1_containers_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/containers/{container_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Retrieve Container
* @description Container retrieve endpoint for getting details of a specific container.
*
* Follows the OpenAI Containers API spec:
* https://platform.openai.com/docs/api-reference/containers
*
* Example:
* ```bash
* curl -X GET "http://localhost:4000/v1/containers/cntr_123" -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*
* Or specify provider via header:
* ```bash
* curl -X GET "http://localhost:4000/v1/containers/cntr_123" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -H "custom-llm-provider: azure"
* ```
*/
get: operations["retrieve_container_v1_containers__container_id__get"];
put?: never;
post?: never;
/**
* Delete Container
* @description Container delete endpoint for deleting a specific container.
*
* Follows the OpenAI Containers API spec:
* https://platform.openai.com/docs/api-reference/containers
*
* Example:
* ```bash
* curl -X DELETE "http://localhost:4000/v1/containers/cntr_123" -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*
* Or specify provider via header:
* ```bash
* curl -X DELETE "http://localhost:4000/v1/containers/cntr_123" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -H "custom-llm-provider: azure"
* ```
*/
delete: operations["delete_container_v1_containers__container_id__delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/containers/{container_id}/files": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Handler Container Id */
get: operations["handler_container_id_v1_containers__container_id__files_get"];
put?: never;
/** Handler Multipart Upload */
post: operations["handler_multipart_upload_v1_containers__container_id__files_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/containers/{container_id}/files/{file_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Handler Container File */
get: operations["handler_container_file_v1_containers__container_id__files__file_id__get"];
put?: never;
post?: never;
/** Handler Container File */
delete: operations["handler_container_file_v1_containers__container_id__files__file_id__delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/containers/{container_id}/files/{file_id}/content": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Handler Binary Content */
get: operations["handler_binary_content_v1_containers__container_id__files__file_id__content_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/decisions": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Ask questions about an input (OpenAI Decisions format)
* @description Judge an `input` against 1 to 128 `questions` and get the answers back in the same order.
*
* Question types are `predicate` (yes/no with `probability`), `choice` (one of the `choices`, with
* `probabilities`), `score` (an index into `levels`, with `probabilities`) and `box` (the image region the
* `instructions` describe, with `confidence`). Requests that break these rules are rejected with a 400 before
* any provider is called. `model` is any decision model in the proxy
* model list. System One decision models accept this format too, LiteLLM translates the request and the
* answers. Streaming is not supported.
*
* [Docs](https://docs.litellm.ai/docs/decisions)
*/
post: operations["decisions_v1_decisions_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/embeddings": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Embeddings
* @description Follows the exact same API spec as `OpenAI's Embeddings API https://platform.openai.com/docs/api-reference/embeddings`
*
* ```bash
* curl -X POST http://localhost:4000/v1/embeddings
* -H "Content-Type: application/json"
* -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* -d '{
* "model": "text-embedding-ada-002",
* "input": "The quick brown fox jumps over the lazy dog"
* }'
* ```
*/
post: operations["embeddings_v1_embeddings_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/evals": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List Evals
* @description List evaluations with pagination.
*
* Model-based routing (for multi-account support):
* - Pass model via header: `x-litellm-model: gpt-4-account-1`
* - Pass model via query: `?model=gpt-4-account-1`
* - Pass model via body: `{"model": "gpt-4-account-1"}`
*
* Example usage:
* ```bash
* curl "http://localhost:4000/v1/evals?limit=10" -H "Authorization: Bearer your-key"
* ```
*
* Returns: ListEvalsResponse with list of evaluations
*/
get: operations["list_evals_v1_evals_get"];
put?: never;
/**
* Create Eval
* @description Create a new evaluation.
*
* Model-based routing (for multi-account support):
* - Pass model via header: `x-litellm-model: gpt-4-account-1`
* - Pass model via query: `?model=gpt-4-account-1`
* - Pass model via body: `{"model": "gpt-4-account-1"}`
*
* Example usage:
* ```bash
* curl -X POST "http://localhost:4000/v1/evals" -H "Authorization: Bearer your-key" -H "Content-Type: application/json" -d '{
* "name": "Test Eval",
* "data_source_config": {"type": "file", "file_id": "file-abc123"},
* "testing_criteria": {"graders": [{"type": "llm_as_judge"}]}
* }'
* ```
*
* Returns: Eval object with id, status, timestamps, etc.
*/
post: operations["create_eval_v1_evals_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/evals/{eval_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Eval
* @description Get a specific evaluation by ID.
*
* Model-based routing (for multi-account support):
* - Pass model via header: `x-litellm-model: gpt-4-account-1`
* - Pass model via query: `?model=gpt-4-account-1`
* - Pass model via body: `{"model": "gpt-4-account-1"}`
*
* Example usage:
* ```bash
* curl "http://localhost:4000/v1/evals/eval_123" -H "Authorization: Bearer your-key"
* ```
*
* Returns: Eval object
*/
get: operations["get_eval_v1_evals__eval_id__get"];
put?: never;
/**
* Update Eval
* @description Update an evaluation.
*
* Model-based routing (for multi-account support):
* - Pass model via header: `x-litellm-model: gpt-4-account-1`
* - Pass model via query: `?model=gpt-4-account-1`
* - Pass model via body: `{"model": "gpt-4-account-1"}`
*
* Example usage:
* ```bash
* curl -X POST "http://localhost:4000/v1/evals/eval_123" -H "Authorization: Bearer your-key" -H "Content-Type: application/json" -d '{"name": "Updated Name"}'
* ```
*
* Returns: Updated Eval object
*/
post: operations["update_eval_v1_evals__eval_id__post"];
/**
* Delete Eval
* @description Delete an evaluation.
*
* Model-based routing (for multi-account support):
* - Pass model via header: `x-litellm-model: gpt-4-account-1`
* - Pass model via query: `?model=gpt-4-account-1`
* - Pass model via body: `{"model": "gpt-4-account-1"}`
*
* Example usage:
* ```bash
* curl -X DELETE "http://localhost:4000/v1/evals/eval_123" -H "Authorization: Bearer your-key"
* ```
*
* Returns: DeleteEvalResponse with deletion confirmation
*/
delete: operations["delete_eval_v1_evals__eval_id__delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/evals/{eval_id}/cancel": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Cancel Eval
* @description Cancel a running evaluation.
*
* Model-based routing (for multi-account support):
* - Pass model via header: `x-litellm-model: gpt-4-account-1`
* - Pass model via query: `?model=gpt-4-account-1`
* - Pass model via body: `{"model": "gpt-4-account-1"}`
*
* Example usage:
* ```bash
* curl -X POST "http://localhost:4000/v1/evals/eval_123/cancel" -H "Authorization: Bearer your-key"
* ```
*
* Returns: CancelEvalResponse with cancellation confirmation
*/
post: operations["cancel_eval_v1_evals__eval_id__cancel_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/evals/{eval_id}/runs": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List Runs
* @description List all runs for an evaluation with pagination.
*
* Model-based routing (for multi-account support):
* - Pass model via header: `x-litellm-model: gpt-4-account-1`
* - Pass model via query: `?model=gpt-4-account-1`
*
* Example usage:
* ```bash
* curl "http://localhost:4000/v1/evals/eval_123/runs?limit=10" -H "Authorization: Bearer your-key"
* ```
*
* Returns: ListRunsResponse with list of runs
*/
get: operations["list_runs_v1_evals__eval_id__runs_get"];
put?: never;
/**
* Create Run
* @description Create a new run for an evaluation.
*
* Model-based routing (for multi-account support):
* - Pass model via header: `x-litellm-model: gpt-4-account-1`
* - Pass model via query: `?model=gpt-4-account-1`
* - Pass model via body: `{"model": "gpt-4-account-1"}`
* - Pass model via completion.model: `{"completion": {"model": "gpt-4-account-1"}}`
*
* Example usage:
* ```bash
* curl -X POST "http://localhost:4000/v1/evals/eval_123/runs" -H "Authorization: Bearer your-key" -H "Content-Type: application/json" -d '{
* "data_source": {"type": "dataset", "dataset_id": "dataset_123"},
* "completion": {"model": "gpt-4", "temperature": 0.7}
* }'
* ```
*
* Returns: Run object with id, status, timestamps, etc.
*/
post: operations["create_run_v1_evals__eval_id__runs_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/evals/{eval_id}/runs/{run_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Run
* @description Get a specific run by ID.
*
* Model-based routing (for multi-account support):
* - Pass model via header: `x-litellm-model: gpt-4-account-1`
* - Pass model via query: `?model=gpt-4-account-1`
*
* Example usage:
* ```bash
* curl "http://localhost:4000/v1/evals/eval_123/runs/run_456" -H "Authorization: Bearer your-key"
* ```
*
* Returns: Run object with full details
*/
get: operations["get_run_v1_evals__eval_id__runs__run_id__get"];
put?: never;
/**
* Cancel Run
* @description Cancel a running run.
*
* Model-based routing (for multi-account support):
* - Pass model via header: `x-litellm-model: gpt-4-account-1`
* - Pass model via query: `?model=gpt-4-account-1`
*
* Example usage:
* ```bash
* curl -X POST "http://localhost:4000/v1/evals/eval_123/runs/run_456/cancel" -H "Authorization: Bearer your-key"
* ```
*
* Returns: CancelRunResponse with cancellation confirmation
*/
post: operations["cancel_run_v1_evals__eval_id__runs__run_id__post"];
/**
* Delete Run
* @description Delete a run.
*
* Model-based routing (for multi-account support):
* - Pass model via header: `x-litellm-model: gpt-4-account-1`
* - Pass model via query: `?model=gpt-4-account-1`
*
* Example usage:
* ```bash
* curl -X DELETE "http://localhost:4000/v1/evals/eval_123/runs/run_456" -H "Authorization: Bearer your-key"
* ```
*
* Returns: RunDeleteResponse with deletion confirmation
*/
delete: operations["delete_run_v1_evals__eval_id__runs__run_id__delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/extraction": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Extract typed fields from a state
* @description Read a `state` text and up to 4 images and get one typed answer per named field in `fields`.
*
* Field types are `text`, `number`, `date`, `time`, `datetime`, `duration`, `list` (set `items`), `estimate`
* (set `min` and `max`) and `forecast` (set `min`, `max` and `horizon`). Each answer carries its `value`, which is
* null when the input does not determine it, plus `confidence`, `p_null`, text `span`s into the state and, for
* estimates, `mean`, `median`, `p10` and `p90`. Images are data URLs or `{content_type, base64}` objects, remote
* URLs are not accepted. An unknown field type, a field name that is not 1 to 64 letters, digits, `_`, `.` or `-`
* starting with a letter or `_`, a remote image URL or more than 4 images is rejected with a 400 before any
* provider is called. Any other top-level key, `stream: true` and OpenAI generation settings such as `temperature`
* included, is rejected with a 400 naming it unless `drop_params` or the deployment's `additional_drop_params`
* drops it. Settings inside `fields` and `options` go to the provider as sent, so a field missing a setting its
* type needs or an empty `state` with no images comes back with the provider's own error. `model` is any
* extraction model in the proxy model list, such as Cloudflare Clairic. The request and the answers keep the
* provider's format.
*/
post: operations["extraction_v1_extraction_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/files": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List Files
* @description Returns information about a specific file. that can be used across - Assistants API, Batch API
* This is the equivalent of GET https://api.openai.com/v1/files/
*
* Supports Identical Params as: https://platform.openai.com/docs/api-reference/files/list
*
* Example Curl
* ```
* curl http://localhost:4000/v1/files -H "Authorization: Bearer $LITELLM_MASTER_KEY"
*
* ```
*/
get: operations["list_files_v1_files_get"];
put?: never;
/**
* Create File
* @description Upload a file that can be used across - Assistants API, Batch API
* This is the equivalent of POST https://api.openai.com/v1/files
*
* Supports Identical Params as: https://platform.openai.com/docs/api-reference/files/create
*
* Example Curl
* ```
* curl http://localhost:4000/v1/files -H "Authorization: Bearer $LITELLM_MASTER_KEY" -F purpose="batch" -F file="@mydata.jsonl"
* -F expires_after[anchor]="created_at" -F expires_after[seconds]=2592000
* ```
*/
post: operations["create_file_v1_files_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/files/{file_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get File
* @description Returns information about a specific file. that can be used across - Assistants API, Batch API
* This is the equivalent of GET https://api.openai.com/v1/files/{file_id}
*
* Supports Identical Params as: https://platform.openai.com/docs/api-reference/files/retrieve
*
* Example Curl
* ```
* curl http://localhost:4000/v1/files/file-abc123 -H "Authorization: Bearer $LITELLM_MASTER_KEY"
*
* ```
*/
get: operations["get_file_v1_files__file_id__get"];
put?: never;
post?: never;
/**
* Delete File
* @description Deletes a specified file. that can be used across - Assistants API, Batch API
* This is the equivalent of DELETE https://api.openai.com/v1/files/{file_id}
*
* Supports Identical Params as: https://platform.openai.com/docs/api-reference/files/delete
*
* Example Curl
* ```
* curl http://localhost:4000/v1/files/file-abc123 -X DELETE -H "Authorization: Bearer $OPENAI_API_KEY"
*
* ```
*/
delete: operations["delete_file_v1_files__file_id__delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/files/{file_id}/content": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get File Content
* @description Returns information about a specific file. that can be used across - Assistants API, Batch API
* This is the equivalent of GET https://api.openai.com/v1/files/{file_id}/content
*
* Supports Identical Params as: https://platform.openai.com/docs/api-reference/files/retrieve-contents
*
* Example Curl
* ```
* curl http://localhost:4000/v1/files/file-abc123/content -H "Authorization: Bearer $LITELLM_MASTER_KEY"
*
* ```
*/
get: operations["get_file_content_v1_files__file_id__content_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/fine_tuning/jobs": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* ✨ (Enterprise) List Fine-Tuning Jobs
* @description Lists fine-tuning jobs for the organization.
* This is the equivalent of GET https://api.openai.com/v1/fine_tuning/jobs
*
* Supported Query Params:
* - `custom_llm_provider`: Name of the LiteLLM provider
* - `after`: Identifier for the last job from the previous pagination request.
* - `limit`: Number of fine-tuning jobs to retrieve (default is 20).
*/
get: operations["list_fine_tuning_jobs_v1_fine_tuning_jobs_get"];
put?: never;
/**
* ✨ (Enterprise) Create Fine-Tuning Job
* @description Creates a fine-tuning job which begins the process of creating a new model from a given dataset.
* This is the equivalent of POST https://api.openai.com/v1/fine_tuning/jobs
*
* Supports Identical Params as: https://platform.openai.com/docs/api-reference/fine-tuning/create
*
* Example Curl:
* ```
* curl http://localhost:4000/v1/fine_tuning/jobs -H "Content-Type: application/json" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -d '{
* "model": "gpt-3.5-turbo",
* "training_file": "file-abc123",
* "hyperparameters": {
* "n_epochs": 4
* }
* }'
* ```
*/
post: operations["create_fine_tuning_job_v1_fine_tuning_jobs_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/fine_tuning/jobs/{fine_tuning_job_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* ✨ (Enterprise) Retrieve Fine-Tuning Job
* @description Retrieves a fine-tuning job.
* This is the equivalent of GET https://api.openai.com/v1/fine_tuning/jobs/{fine_tuning_job_id}
*
* Supported Query Params:
* - `custom_llm_provider`: Name of the LiteLLM provider
* - `fine_tuning_job_id`: The ID of the fine-tuning job to retrieve.
*/
get: operations["retrieve_fine_tuning_job_v1_fine_tuning_jobs__fine_tuning_job_id__get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/fine_tuning/jobs/{fine_tuning_job_id}/cancel": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* ✨ (Enterprise) Cancel Fine-Tuning Jobs
* @description Cancel a fine-tuning job.
*
* This is the equivalent of POST https://api.openai.com/v1/fine_tuning/jobs/{fine_tuning_job_id}/cancel
*
* Supported Query Params:
* - `custom_llm_provider`: Name of the LiteLLM provider
* - `fine_tuning_job_id`: The ID of the fine-tuning job to cancel.
*/
post: operations["cancel_fine_tuning_job_v1_fine_tuning_jobs__fine_tuning_job_id__cancel_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/images/edits": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Image Edit Api
* @description Follows the OpenAI Images API spec: https://platform.openai.com/docs/api-reference/images/create
*
* ```bash
* curl -s -D >(grep -i x-request-id >&2) -o >(jq -r '.data[0].b64_json' | base64 --decode > gift-basket.png) -X POST "http://localhost:4000/v1/images/edits" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -F "model=gpt-image-1" -F "image[][email protected]" -F 'prompt=Create a studio ghibli image of this'
* ```
*/
post: operations["image_edit_api_v1_images_edits_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/images/generations": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/** Image Generation */
post: operations["image_generation_v1_images_generations_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/indexes": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Index List
* @description List all vector store indexes. Proxy admin only.
*
* ```bash
* curl -L -X GET 'http://0.0.0.0:4000/v1/indexes' -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*/
get: operations["index_list_v1_indexes_get"];
put?: never;
/**
* Index Create
* @description Create an index. Just writes the index to the database.
*
* ```bash
* curl -L -X POST 'http://0.0.0.0:4000/v1/indexes' -H 'Content-Type: application/json' -H "Authorization: Bearer $LITELLM_MASTER_KEY" -d '{
* "index_name": "dall-e-3",
* "litellm_params": {
* "vector_store_index": "real-index-name",
* "vector_store_name": "azure-ai-search"
* }
* }'
* ```
*/
post: operations["index_create_v1_indexes_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/live/sessions": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* WebSocket: live_sessions_websocket_endpoint
* @description WebSocket connection endpoint
*/
get: operations["websocket_live_sessions_websocket_endpoint_get_2"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/logs": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/** Ingest Otlp Traces */
post: operations["ingest_otlp_traces_v1_logs_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/mcp/access_groups": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Mcp Access Groups
* @description Get all available MCP access groups from the database AND config
*/
get: operations["get_mcp_access_groups_v1_mcp_access_groups_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/mcp/discover": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Discover Mcp Servers
* @description Returns a curated list of well-known MCP servers for discovery UI
*/
get: operations["discover_mcp_servers_v1_mcp_discover_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/mcp/make_public": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Make Mcp Servers Public
* @description Allows making MCP servers public for AI Hub
*/
post: operations["make_mcp_servers_public_v1_mcp_make_public_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/mcp/network/client-ip": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Client Ip
* @description Returns the caller's IP address as seen by the proxy.
*/
get: operations["get_client_ip_v1_mcp_network_client_ip_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/mcp/openapi-registry": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Openapi Registry
* @description Returns well-known OpenAPI APIs with OAuth 2.0 metadata for the OpenAPI MCP picker
*/
get: operations["get_openapi_registry_v1_mcp_openapi_registry_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/mcp/registry.json": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Mcp Registry
* @description MCP registry endpoint. Spec: https://github.com/modelcontextprotocol/registry
*/
get: operations["get_mcp_registry_v1_mcp_registry_json_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/mcp/server": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Fetch All Mcp Servers
* @description Returns the mcp server list with associated teams
*/
get: operations["fetch_all_mcp_servers_v1_mcp_server_get"];
/**
* Edit Mcp Server
* @description Allows deleting mcp serves in the db
*/
put: operations["edit_mcp_server_v1_mcp_server_put"];
/**
* Add Mcp Server
* @description Allows creation of mcp servers
*/
post: operations["add_mcp_server_v1_mcp_server_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/mcp/server/health": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Health Check Servers
* @description Health check for MCP servers
*/
get: operations["health_check_servers_v1_mcp_server_health_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/mcp/server/import": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Import Mcp Servers
* @description Bulk-import MCP connectors from Anthropic mcpServers or mcp_servers JSON
*/
post: operations["import_mcp_servers_v1_mcp_server_import_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/mcp/server/oauth/session": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Add Session Mcp Server
* @description Temporarily cache an MCP server in memory without writing to the database
*/
post: operations["add_session_mcp_server_v1_mcp_server_oauth_session_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/mcp/server/register": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Register Mcp Server
* @description Submit a new MCP server for admin review (non-admin users). Mirrors POST /guardrails/register.
*/
post: operations["register_mcp_server_v1_mcp_server_register_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/mcp/server/submissions": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Mcp Server Submissions
* @description Returns all MCP servers submitted by non-admin users (admin review queue). Mirrors GET /guardrails/submissions.
*/
get: operations["get_mcp_server_submissions_v1_mcp_server_submissions_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/mcp/server/{server_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Fetch Mcp Server
* @description Returns the mcp server info
*/
get: operations["fetch_mcp_server_v1_mcp_server__server_id__get"];
put?: never;
post?: never;
/**
* Remove Mcp Server
* @description Allows deleting mcp serves in the db
*/
delete: operations["remove_mcp_server_v1_mcp_server__server_id__delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/mcp/server/{server_id}/approve": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
/**
* Approve Mcp Server Submission
* @description Approve a pending MCP server submission (admin only). Mirrors PUT /guardrails/{id}/approve.
*/
put: operations["approve_mcp_server_submission_v1_mcp_server__server_id__approve_put"];
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/mcp/server/{server_id}/oauth-user-credential": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Store Mcp Oauth User Credential
* @description Store the calling user's OAuth2 token for an OpenAPI MCP server
*/
post: operations["store_mcp_oauth_user_credential_v1_mcp_server__server_id__oauth_user_credential_post"];
/**
* Delete Mcp Oauth User Credential
* @description Revoke the calling user's stored OAuth2 token for an MCP server. A proxy admin may pass user_id to revoke another user's stored token.
*/
delete: operations["delete_mcp_oauth_user_credential_v1_mcp_server__server_id__oauth_user_credential_delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/mcp/server/{server_id}/oauth-user-credential/status": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Mcp Oauth User Credential Status
* @description Check whether the calling user has a stored OAuth2 credential for this MCP server
*/
get: operations["get_mcp_oauth_user_credential_status_v1_mcp_server__server_id__oauth_user_credential_status_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/mcp/server/{server_id}/pin": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Pin Mcp Server Tools
* @description Pin the server's current upstream tool list, descriptions and input schemas (admin only). tools/list serves the pinned catalog from now on and an upstream change raises an mcp_pinned_tools_changed alert.
*/
post: operations["pin_mcp_server_tools_v1_mcp_server__server_id__pin_post"];
/**
* Unpin Mcp Server Tools
* @description Unpin the server's tool list (admin only); tools/list serves the live upstream catalog again.
*/
delete: operations["unpin_mcp_server_tools_v1_mcp_server__server_id__pin_delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/mcp/server/{server_id}/reject": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
/**
* Reject Mcp Server Submission
* @description Reject a pending MCP server submission (admin only). Mirrors PUT /guardrails/{id}/reject.
*/
put: operations["reject_mcp_server_submission_v1_mcp_server__server_id__reject_put"];
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/mcp/server/{server_id}/user-credential": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Store Mcp User Credential
* @description Store or update the calling user's API key for a BYOK MCP server
*/
post: operations["store_mcp_user_credential_v1_mcp_server__server_id__user_credential_post"];
/**
* Delete Mcp User Credential
* @description Delete the calling user's stored API key for a BYOK MCP server. A proxy admin may pass user_id to revoke another user's stored key.
*/
delete: operations["delete_mcp_user_credential_v1_mcp_server__server_id__user_credential_delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/mcp/server/{server_id}/user-credentials": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List Mcp Server User Credentials
* @description List every user's stored BYOK or OAuth2 credential for an MCP server (admin only, no secrets)
*/
get: operations["list_mcp_server_user_credentials_v1_mcp_server__server_id__user_credentials_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/mcp/server/{server_id}/user-env-vars": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Mcp User Env Vars
* @description Return the calling user's per-user MCP env var status for this server.
*/
get: operations["get_mcp_user_env_vars_v1_mcp_server__server_id__user_env_vars_get"];
put?: never;
/**
* Store Mcp User Env Vars
* @description Store the calling user's per-user MCP env var values for this server. Submitted values are merged over any previously stored values, so you only send the fields you want to set or change; a variable omitted (or sent empty) keeps its stored value. Use DELETE to clear all stored values.
*/
post: operations["store_mcp_user_env_vars_v1_mcp_server__server_id__user_env_vars_post"];
/**
* Clear Mcp User Env Vars
* @description Clear the calling user's per-user MCP env var values for this server.
*/
delete: operations["clear_mcp_user_env_vars_v1_mcp_server__server_id__user_env_vars_delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/mcp/sessions": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Mcp Gateway Sessions
* @description Live stateful MCP gateway sessions on this proxy worker, grouped by AI client and by user.
*/
get: operations["get_mcp_gateway_sessions_v1_mcp_sessions_get"];
put?: never;
post?: never;
/**
* Delete Mcp Gateway Sessions
* @description Force-close live stateful MCP gateway sessions on this proxy worker, selected by session id prefix and/or by the LiteLLM user that opened them (proxy admin only).
*/
delete: operations["delete_mcp_gateway_sessions_v1_mcp_sessions_delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/mcp/tools": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Mcp Tools
* @description Get all MCP tools available for the current key, including those from access groups
*/
get: operations["get_mcp_tools_v1_mcp_tools_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/mcp/toolset": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Fetch Mcp Toolsets
* @description List MCP toolsets accessible to the calling key
*/
get: operations["fetch_mcp_toolsets_v1_mcp_toolset_get"];
/**
* Edit Mcp Toolset
* @description Update an existing MCP toolset (admin only)
*/
put: operations["edit_mcp_toolset_v1_mcp_toolset_put"];
/**
* Add Mcp Toolset
* @description Create a new MCP toolset (admin only)
*/
post: operations["add_mcp_toolset_v1_mcp_toolset_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/mcp/toolset/{toolset_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Fetch Mcp Toolset
* @description Get a specific MCP toolset by ID
*/
get: operations["fetch_mcp_toolset_v1_mcp_toolset__toolset_id__get"];
put?: never;
post?: never;
/**
* Remove Mcp Toolset
* @description Delete an MCP toolset (admin only)
*/
delete: operations["remove_mcp_toolset_v1_mcp_toolset__toolset_id__delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/mcp/user-credentials": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List Mcp User Credentials
* @description List all OAuth2 MCP credentials stored for the calling user
*/
get: operations["list_mcp_user_credentials_v1_mcp_user_credentials_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/mcp/user-env-vars/status": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List Mcp User Env Var Status
* @description Per-user MCP env var status across every server the user can access. Used by the dashboard to highlight servers with missing per-user vars.
*/
get: operations["list_mcp_user_env_var_status_v1_mcp_user_env_vars_status_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/memory": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List Memory
* @description List memory entries visible to the caller.
*/
get: operations["list_memory_v1_memory_get"];
put?: never;
/**
* Create Memory
* @description Create a new memory entry for the caller (or, for admins, any scope).
*/
post: operations["create_memory_v1_memory_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/memory/{key}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Memory
* @description Get a single memory entry by key, scoped to the caller.
*/
get: operations["get_memory_v1_memory__key__get"];
/**
* Upsert Memory
* @description Upsert a memory entry by key within the caller's scope.
*
* If no row exists for (key, caller.user_id, caller.team_id), create one.
* If one exists, update the value/metadata fields that were provided.
*/
put: operations["upsert_memory_v1_memory__key__put"];
post?: never;
/**
* Delete Memory
* @description Delete a memory entry by key, scoped to the caller.
*/
delete: operations["delete_memory_v1_memory__key__delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/messages": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Anthropic Response
* @description Use `{PROXY_BASE_URL}/anthropic/v1/messages` instead - [Docs](https://docs.litellm.ai/docs/pass_through/anthropic_completion).
*
* This was a BETA endpoint that calls 100+ LLMs in the anthropic format.
*/
post: operations["anthropic_response_v1_messages_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/messages/count_tokens": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Count Tokens
* @description Count tokens for Anthropic Messages API format.
*
* This endpoint follows the Anthropic Messages API token counting specification.
* It accepts the same parameters as the /v1/messages endpoint but returns
* token counts instead of generating a response.
*
* Example usage:
* ```
* curl -X POST "http://localhost:4000/v1/messages/count_tokens?beta=true" -H "Content-Type: application/json" -H "Authorization: Bearer your-key" -d '{
* "model": "claude-3-sonnet-20240229",
* "messages": [{"role": "user", "content": "Hello Claude!"}]
* }'
* ```
*
* Returns: {"input_tokens": <number>}
*/
post: operations["count_tokens_v1_messages_count_tokens_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/model/deprecations": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Model Deprecations
* @description List models with known deprecation/sunset dates, bucketed by urgency.
*
* Reads `deprecation_date` metadata from `model_prices_and_context_window.json`
* (and any per-deployment `model_info.deprecation_date` overrides) for the
* models configured on this proxy. An entry also names its `successor_model`
* when the deployment sets `model_info.successor_model`.
*
* Parameters:
* warn_within_days: Window (in days) used to bucket "imminent" models,
* 30 by default.
*
* Returns:
* A payload with three lists of `ModelDeprecationInfo` entries:
*
* - `deprecated`: deprecation date is in the past, so these requests may
* fail at any time.
* - `imminent`: deprecation date is within `warn_within_days` from today.
* - `upcoming`: deprecation date is further out.
*
* Example:
* ```shell
* curl -X GET 'http://localhost:4000/model/deprecations' \
* -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*/
get: operations["model_deprecations_v1_model_deprecations_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/model/info": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Model Info V1
* @description Provides more info about each model in /models, including config.yaml descriptions (except api key and api base)
*
* Parameters:
* litellm_model_id: Optional[str] = None (this is the value of `x-litellm-model-id` returned in response headers)
*
* - When litellm_model_id is passed, it will return the info for that specific model
* - When litellm_model_id is not passed, it will return the info for all models
* - include_team_models: When true, filter to deployments the caller can use (same as /v2/model/info).
* - teamId: Filter to models accessible by the given team.
* - healthy_only: When true, hide models whose backing deployments are all marked
* unhealthy by background health checks, matching `/v1/models?healthy_only=true`.
* Set `general_settings.model_list_healthy_only: true` to apply this to every
* caller without the query parameter. Requires `background_health_checks: true`,
* plus either `model_list_healthy_only` or `enable_health_check_routing` to keep
* deployment health state cached; without health state the listing is returned
* unfiltered (fail open). Ignored when `litellm_model_id` is passed, since that
* is a direct lookup of one deployment rather than a listing. Hiding is
* presentation-only: a hidden model can still be called directly.
*
* Each model in the list response includes `model_info.access_via_team_ids` and
* `model_info.direct_access` when the proxy database is connected.
*
* Returns:
* A JSON response whose `data` list holds one entry per model.
*
* Example Response:
* ```json
* {
* "data": [
* {
* "model_name": "fake-openai-endpoint",
* "litellm_params": {
* "api_base": "https://exampleopenaiendpoint-production.up.railway.app/",
* "model": "openai/fake"
* },
* "model_info": {
* "id": "112f74fab24a7a5245d2ced3536dd8f5f9192c57ee6e332af0f0512e08bed5af",
* "db_model": false
* }
* }
* ]
* }
*
* ```
*/
get: operations["model_info_v1_v1_model_info_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/models": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Model List
* @description Use `/model/info` - to get detailed model information, example - pricing, mode, etc.
*
* This is just for compatibility with openai projects like aider.
*
* Query Parameters:
* - client_version: Sent by Codex CLI (`?client_version=0.159.3`) when it fetches a
* provider's model catalog. When present, the response is Codex's own
* catalog shape (`{"models": [...]}`) built from the same listing, with
* each model's `model_info.service_tiers` as its service tiers; absent,
* the OpenAI shape below is returned
* - include_metadata: Include additional metadata in the response with fallback information
* - fallback_type: Type of fallbacks to include ("general", "context_window", "content_policy")
* Defaults to "general" when include_metadata=true
* - scope: Optional scope parameter. Currently only accepts "expand".
* When scope=expand is passed, proxy admins, team admins, and org admins
* will receive all proxy models as if they are a proxy admin.
* - healthy_only: When true, hide models whose backing deployments are all marked
* unhealthy by background health checks. Set
* `general_settings.model_list_healthy_only: true` to apply this
* to every caller without the query parameter. Requires
* `background_health_checks: true` in general_settings, plus
* either `model_list_healthy_only` or `enable_health_check_routing`
* to keep deployment health state cached; without health state
* the listing is returned unfiltered (fail open).
* Models expanded from wildcard routes (e.g. `openai/*`) are not
* filtered, and nothing is hidden when `allowed_fails_policy` is
* configured (cooldown remains the sole exclusion mechanism).
* Hiding is presentation-only: a hidden model can still be
* called directly.
*/
get: operations["model_list_v1_models_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/models/{model_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Model Info
* @description Retrieve information about a specific model accessible to your API key.
*
* Returns model details only if the model is available to your API key/team.
* Returns 404 if the model doesn't exist or is not accessible.
*
* Follows OpenAI API specification for individual model retrieval.
* https://platform.openai.com/docs/api-reference/models/retrieve
*
* Query parameters mirror `/v1/models` so the same caller context (team
* scoping, health filtering, paused deployments) drives both endpoints; the
* listing's public id must resolve to the same internal deployment here.
*/
get: operations["model_info_v1_models__model_id__get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/moderations": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Moderations
* @description The moderations endpoint is a tool you can use to check whether content complies with an LLM Providers policies.
* Quick Start
* ```
* curl --location 'http://0.0.0.0:4000/moderations' --header 'Content-Type: application/json' --header "Authorization: Bearer $LITELLM_MASTER_KEY" --data '{"input": "Sample text goes here", "model": "text-moderation-stable"}'
* ```
*/
post: operations["moderations_v1_moderations_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/ocr": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Ocr
* @description OCR endpoint for extracting text from documents and images.
*
* Supports two input modes:
*
* **1. JSON body** (Mistral OCR API compatible):
* ```bash
* curl -X POST "http://localhost:4000/v1/ocr" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -H "Content-Type: application/json" -d '{
* "model": "mistral-ocr",
* "document": {
* "type": "document_url",
* "document_url": "https://arxiv.org/pdf/2201.04234"
* }
* }'
* ```
*
* **2. Multipart form file upload**:
* ```bash
* curl -X POST "http://localhost:4000/v1/ocr" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -F "model=mistral-ocr" -F "[email protected]"
* ```
*
* Response format is normalized to the LiteLLM OCR schema by default. Providers
* that support it (Azure Document Intelligence) can return their own payload
* instead, with cost tracking unchanged, via `x-req-format: native` (or
* `"req_format": "native"` in the body).
*/
post: operations["ocr_v1_ocr_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/rag/ingest": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Rag Ingest
* @description RAG Ingest endpoint - all-in-one document ingestion pipeline.
*
* Supports form upload (for files) or JSON body (for URLs).
*
* ## Form upload (for files):
* ```bash
* curl -X POST "http://localhost:4000/v1/rag/ingest" \
* -H "Authorization: Bearer $LITELLM_MASTER_KEY" \
* -F file="@document.pdf" \
* -F 'ingest_options={"vector_store": {"custom_llm_provider": "openai"}}'
* ```
*
* ## JSON body (for URLs):
* ```bash
* curl -X POST "http://localhost:4000/v1/rag/ingest" \
* -H "Authorization: Bearer $LITELLM_MASTER_KEY" \
* -H "Content-Type: application/json" \
* -d '{
* "file_url": "https://example.com/document.pdf",
* "ingest_options": {"vector_store": {"custom_llm_provider": "openai"}}
* }'
* ```
*
* ## Bedrock:
* ```bash
* curl -X POST "http://localhost:4000/v1/rag/ingest" \
* -H "Authorization: Bearer $LITELLM_MASTER_KEY" \
* -F file="@document.pdf" \
* -F 'ingest_options={"vector_store": {"custom_llm_provider": "bedrock"}}'
* ```
*/
post: operations["rag_ingest_v1_rag_ingest_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/rag/query": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Rag Query
* @description RAG Query endpoint - search vector store, optionally rerank, and generate LLM response.
*
* This endpoint:
* 1. Extracts the query from the last user message
* 2. Searches the vector store for relevant context
* 3. Optionally reranks the results
* 4. Generates an LLM response with the retrieved context
*
* ## Example Request:
* ```bash
* curl -X POST "http://localhost:4000/v1/rag/query" \
* -H "Authorization: Bearer $LITELLM_MASTER_KEY" \
* -H "Content-Type: application/json" \
* -d '{
* "model": "gpt-4o-mini",
* "messages": [{"role": "user", "content": "What is LiteLLM?"}],
* "retrieval_config": {
* "vector_store_id": "vs_abc123",
* "custom_llm_provider": "openai",
* "top_k": 5
* }
* }'
* ```
*
* ## With Reranking:
* ```bash
* curl -X POST "http://localhost:4000/v1/rag/query" \
* -H "Authorization: Bearer $LITELLM_MASTER_KEY" \
* -H "Content-Type: application/json" \
* -d '{
* "model": "gpt-4o-mini",
* "messages": [{"role": "user", "content": "What is LiteLLM?"}],
* "retrieval_config": {
* "vector_store_id": "vs_abc123",
* "custom_llm_provider": "openai",
* "top_k": 10
* },
* "rerank": {
* "enabled": true,
* "model": "cohere/rerank-english-v3.0",
* "top_n": 3
* }
* }'
* ```
*/
post: operations["rag_query_v1_rag_query_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/realtime": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* WebSocket: realtime_websocket_endpoint
* @description WebSocket connection endpoint
*/
get: operations["websocket_realtime_websocket_endpoint_get_2"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/realtime/calls": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/** Proxy Realtime Calls */
post: operations["proxy_realtime_calls_v1_realtime_calls_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/realtime/client_secrets": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/** Create Realtime Client Secret */
post: operations["create_realtime_client_secret_v1_realtime_client_secrets_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/realtime/transcription_sessions": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Create Realtime Transcription Session
* @description Create an ephemeral Realtime transcription session
* (POST /v1/realtime/transcription_sessions) for the WebRTC/WebSocket flow.
*
* Mirrors the client_secrets route but targets the transcription_sessions
* endpoint and encrypts the ephemeral key returned under `client_secret.value`.
*/
post: operations["create_realtime_transcription_session_v1_realtime_transcription_sessions_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/rerank": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/** Rerank */
post: operations["rerank_v1_rerank_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/responses": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* WebSocket: responses_websocket_endpoint
* @description WebSocket connection endpoint
*/
get: operations["websocket_responses_websocket_endpoint_get_2"];
put?: never;
/**
* Responses Api
* @description Follows the OpenAI Responses API spec: https://platform.openai.com/docs/api-reference/responses
*
* Supports background mode with polling_via_cache for partial response retrieval.
* When background=true and polling_via_cache is enabled, returns a polling_id immediately
* and streams the response in the background, updating Redis cache.
*
* ```bash
* # Normal request
* curl -X POST http://localhost:4000/v1/responses -H "Content-Type: application/json" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -d '{
* "model": "gpt-4o",
* "input": "Tell me about AI"
* }'
*
* # Background request with polling
* curl -X POST http://localhost:4000/v1/responses -H "Content-Type: application/json" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -d '{
* "model": "gpt-4o",
* "input": "Tell me about AI",
* "background": true
* }'
* ```
*/
post: operations["responses_api_v1_responses_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/responses/compact": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Compact Response
* @description Compact a response by running a compaction pass over a conversation.
*
* Returns encrypted, opaque items that can be used to reduce context size.
*
* Follows the OpenAI Responses API spec: https://platform.openai.com/docs/api-reference/responses/compact
*
* ```bash
* curl -X POST http://localhost:4000/v1/responses/compact -H "Content-Type: application/json" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -d '{
* "model": "gpt-4o",
* "input": [{"role": "user", "content": "Hello"}]
* }'
* ```
*/
post: operations["compact_response_v1_responses_compact_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/responses/input_tokens": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Responses Input Tokens
* @description Count the input tokens of a Responses API request without calling the model.
*
* Follows the OpenAI Responses API spec: https://platform.openai.com/docs/api-reference/responses/input-tokens
*
* ```bash
* curl -X POST http://localhost:4000/v1/responses/input_tokens -H "Content-Type: application/json" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -d '{
* "model": "gpt-4o",
* "input": "Hello, how are you?"
* }'
* ```
*
* Returns: `{"object": "response.input_tokens", "input_tokens": <count>}`
*/
post: operations["responses_input_tokens_v1_responses_input_tokens_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/responses/{response_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Response
* @description Get a response by ID.
*
* Supports both:
* - Polling IDs (litellm_poll_*): Returns cumulative cached content from background responses
* - Provider response IDs: Passes through to provider API
*
* Follows the OpenAI Responses API spec: https://platform.openai.com/docs/api-reference/responses/get
*
* ```bash
* # Get polling response
* curl -X GET http://localhost:4000/v1/responses/litellm_poll_abc123 -H "Authorization: Bearer $LITELLM_MASTER_KEY"
*
* # Get provider response
* curl -X GET http://localhost:4000/v1/responses/resp_abc123 -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*/
get: operations["get_response_v1_responses__response_id__get"];
put?: never;
post?: never;
/**
* Delete Response
* @description Delete a response by ID.
*
* Supports both:
* - Polling IDs (litellm_poll_*): Deletes from Redis cache
* - Provider response IDs: Passes through to provider API
*
* Follows the OpenAI Responses API spec: https://platform.openai.com/docs/api-reference/responses/delete
*
* ```bash
* curl -X DELETE http://localhost:4000/v1/responses/resp_abc123 -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*/
delete: operations["delete_response_v1_responses__response_id__delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/responses/{response_id}/cancel": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Cancel Response
* @description Cancel a response by ID.
*
* Supports both:
* - Polling IDs (litellm_poll_*): Cancels background response and updates status in Redis
* - Provider response IDs: Passes through to provider API
*
* Follows the OpenAI Responses API spec: https://platform.openai.com/docs/api-reference/responses/cancel
*
* ```bash
* # Cancel polling response
* curl -X POST http://localhost:4000/v1/responses/litellm_poll_abc123/cancel -H "Authorization: Bearer $LITELLM_MASTER_KEY"
*
* # Cancel provider response
* curl -X POST http://localhost:4000/v1/responses/resp_abc123/cancel -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*/
post: operations["cancel_response_v1_responses__response_id__cancel_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/responses/{response_id}/input_items": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Response Input Items
* @description List input items for a response.
*/
get: operations["get_response_input_items_v1_responses__response_id__input_items_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/rust_control_plane/logs": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Ingest Callback Logs
* @description Replay a batch of finished logging payloads through the callback fan-out.
*
* Admin-only: the payloads write spend logs and trigger every callback, so this
* is a trusted internal route, not a public surface.
*/
post: operations["ingest_callback_logs_v1_rust_control_plane_logs_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/search": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Search
* @description Search endpoint for performing web searches.
*
* Follows the Perplexity Search API spec:
* https://docs.perplexity.ai/api-reference/search-post
*
* The search_tool_name can be passed either:
* 1. In the URL path: /v1/search/{search_tool_name}
* 2. In the request body: {"search_tool_name": "..."}
*
* Example with search_tool_name in URL (recommended - keeps body Perplexity-compatible):
* ```bash
* curl -X POST "http://localhost:4000/v1/search/litellm-search" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -H "Content-Type: application/json" -d '{
* "query": "latest AI developments 2024",
* "max_results": 5,
* "search_domain_filter": ["arxiv.org", "nature.com"],
* "country": "US"
* }'
* ```
*
* Example with search_tool_name in body:
* ```bash
* curl -X POST "http://localhost:4000/v1/search" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -H "Content-Type: application/json" -d '{
* "search_tool_name": "litellm-search",
* "query": "latest AI developments 2024",
* "max_results": 5,
* "search_domain_filter": ["arxiv.org", "nature.com"],
* "country": "US"
* }'
* ```
*
* Request Body Parameters (when search_tool_name not in URL):
* - search_tool_name (str, required if not in URL): Name of the search tool configured in router
* - query (str or list[str], required): Search query
* - max_results (int, optional): Maximum number of results (1-20), default 10
* - search_domain_filter (list[str], optional): List of domains to filter (max 20)
* - max_tokens_per_page (int, optional): Max tokens per page, default 1024
* - country (str, optional): Country code filter (e.g., 'US', 'GB', 'DE')
*
* When using URL path parameter, only Perplexity-compatible parameters are needed in body:
* - query (str or list[str], required): Search query
* - max_results (int, optional): Maximum number of results (1-20), default 10
* - search_domain_filter (list[str], optional): List of domains to filter (max 20)
* - max_tokens_per_page (int, optional): Max tokens per page, default 1024
* - country (str, optional): Country code filter (e.g., 'US', 'GB', 'DE')
*
* Response follows Perplexity Search API format:
* ```json
* {
* "object": "search",
* "results": [
* {
* "title": "Result title",
* "url": "https://example.com",
* "snippet": "Result snippet...",
* "date": "2024-01-01",
* "last_updated": "2024-01-01"
* }
* ]
* }
* ```
*/
post: operations["search_v1_search_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/search/tools": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List Search Tools
* @description List all available search tools configured in the router.
*
* This endpoint returns the search tools that are currently loaded and available
* for use with the /v1/search endpoint.
*
* Example:
* ```bash
* curl -X GET "http://localhost:4000/v1/search/tools" -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*
* Response:
* ```json
* {
* "object": "list",
* "data": [
* {
* "search_tool_name": "litellm-search",
* "search_provider": "perplexity",
* "description": "Perplexity search tool"
* }
* ]
* }
* ```
*/
get: operations["list_search_tools_v1_search_tools_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/search/{search_tool_name}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Search
* @description Search endpoint for performing web searches.
*
* Follows the Perplexity Search API spec:
* https://docs.perplexity.ai/api-reference/search-post
*
* The search_tool_name can be passed either:
* 1. In the URL path: /v1/search/{search_tool_name}
* 2. In the request body: {"search_tool_name": "..."}
*
* Example with search_tool_name in URL (recommended - keeps body Perplexity-compatible):
* ```bash
* curl -X POST "http://localhost:4000/v1/search/litellm-search" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -H "Content-Type: application/json" -d '{
* "query": "latest AI developments 2024",
* "max_results": 5,
* "search_domain_filter": ["arxiv.org", "nature.com"],
* "country": "US"
* }'
* ```
*
* Example with search_tool_name in body:
* ```bash
* curl -X POST "http://localhost:4000/v1/search" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -H "Content-Type: application/json" -d '{
* "search_tool_name": "litellm-search",
* "query": "latest AI developments 2024",
* "max_results": 5,
* "search_domain_filter": ["arxiv.org", "nature.com"],
* "country": "US"
* }'
* ```
*
* Request Body Parameters (when search_tool_name not in URL):
* - search_tool_name (str, required if not in URL): Name of the search tool configured in router
* - query (str or list[str], required): Search query
* - max_results (int, optional): Maximum number of results (1-20), default 10
* - search_domain_filter (list[str], optional): List of domains to filter (max 20)
* - max_tokens_per_page (int, optional): Max tokens per page, default 1024
* - country (str, optional): Country code filter (e.g., 'US', 'GB', 'DE')
*
* When using URL path parameter, only Perplexity-compatible parameters are needed in body:
* - query (str or list[str], required): Search query
* - max_results (int, optional): Maximum number of results (1-20), default 10
* - search_domain_filter (list[str], optional): List of domains to filter (max 20)
* - max_tokens_per_page (int, optional): Max tokens per page, default 1024
* - country (str, optional): Country code filter (e.g., 'US', 'GB', 'DE')
*
* Response follows Perplexity Search API format:
* ```json
* {
* "object": "search",
* "results": [
* {
* "title": "Result title",
* "url": "https://example.com",
* "snippet": "Result snippet...",
* "date": "2024-01-01",
* "last_updated": "2024-01-01"
* }
* ]
* }
* ```
*/
post: operations["search_v1_search__search_tool_name__post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/skills": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List Skills
* @description List skills on Anthropic.
*
* Requires `?beta=true` query parameter.
*
* Model-based routing (for multi-account support):
* - Pass model via header: `x-litellm-model: claude-account-1`
* - Pass model via query: `?model=claude-account-1`
* - Pass model via body: `{"model": "claude-account-1"}`
*
* Example usage:
* ```bash
* # Basic usage
* curl "http://localhost:4000/v1/skills?beta=true&limit=10" -H "Authorization: Bearer your-key"
*
* # With model-based routing
* curl "http://localhost:4000/v1/skills?beta=true&limit=10" -H "Authorization: Bearer your-key" -H "x-litellm-model: claude-account-1"
* ```
*
* Pass `?custom_llm_provider=litellm_proxy&query=<task>` to rank the LiteLLM-hosted skills you can
* access by semantic similarity instead of paging through the whole registry:
* ```bash
* curl "http://localhost:4000/v1/skills?custom_llm_provider=litellm_proxy&query=summarize+a+pdf&top_k=5" -H "Authorization: Bearer your-key"
* ```
*
* Returns: ListSkillsResponse with list of skills
*/
get: operations["list_skills_v1_skills_get"];
put?: never;
/**
* Create Skill
* @description Create a new skill on Anthropic.
*
* Requires `?beta=true` query parameter.
*
* Model-based routing (for multi-account support):
* - Pass model via header: `x-litellm-model: claude-account-1`
* - Pass model via query: `?model=claude-account-1`
* - Pass model via form field: `model=claude-account-1`
*
* Example usage:
* ```bash
* # Basic usage
* curl -X POST "http://localhost:4000/v1/skills?beta=true" -H "Content-Type: multipart/form-data" -H "Authorization: Bearer your-key" -F "display_title=My Skill" -F "files[][email protected]"
*
* # With model-based routing
* curl -X POST "http://localhost:4000/v1/skills?beta=true" -H "Content-Type: multipart/form-data" -H "Authorization: Bearer your-key" -H "x-litellm-model: claude-account-1" -F "display_title=My Skill" -F "files[][email protected]"
* ```
*
* Returns: Skill object with id, display_title, etc.
*/
post: operations["create_skill_v1_skills_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/skills/{skill_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Skill
* @description Get a specific skill by ID from Anthropic.
*
* Requires `?beta=true` query parameter.
*
* Model-based routing (for multi-account support):
* - Pass model via header: `x-litellm-model: claude-account-1`
* - Pass model via query: `?model=claude-account-1`
* - Pass model via body: `{"model": "claude-account-1"}`
*
* Example usage:
* ```bash
* # Basic usage
* curl "http://localhost:4000/v1/skills/skill_123?beta=true" -H "Authorization: Bearer your-key"
*
* # With model-based routing
* curl "http://localhost:4000/v1/skills/skill_123?beta=true" -H "Authorization: Bearer your-key" -H "x-litellm-model: claude-account-1"
* ```
*
* Returns: Skill object
*/
get: operations["get_skill_v1_skills__skill_id__get"];
put?: never;
post?: never;
/**
* Delete Skill
* @description Delete a skill by ID from Anthropic.
*
* Requires `?beta=true` query parameter.
*
* Note: Anthropic does not allow deleting skills with existing versions.
*
* Model-based routing (for multi-account support):
* - Pass model via header: `x-litellm-model: claude-account-1`
* - Pass model via query: `?model=claude-account-1`
* - Pass model via body: `{"model": "claude-account-1"}`
*
* Example usage:
* ```bash
* # Basic usage
* curl -X DELETE "http://localhost:4000/v1/skills/skill_123?beta=true" -H "Authorization: Bearer your-key"
*
* # With model-based routing
* curl -X DELETE "http://localhost:4000/v1/skills/skill_123?beta=true" -H "Authorization: Bearer your-key" -H "x-litellm-model: claude-account-1"
* ```
*
* Returns: DeleteSkillResponse with type="skill_deleted"
*/
delete: operations["delete_skill_v1_skills__skill_id__delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/skills/{skill_id}/archive": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Agent Skills Archive
* @description Stored skill upload, repacked so SKILL.md sits at the archive root.
*/
get: operations["agent_skills_archive_v1_skills__skill_id__archive_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/systemone": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Ask named questions about a state (System One format)
* @description Judge a `state` against 1 to 128 named `questions` and get one calibrated answer per question name.
*
* Question types are `noul` (probability the answer is yes), `choice` (one label out of `criteria`, with
* `confidence` and `probabilities`), `score` (an index into the `criteria` levels, with `confidence`,
* `legend` and `probabilities`) and `box` (the image region the `instructions` describe, with `confidence`).
* TypeSafe models also take images in `assets`, each one referenced by its key in `state`. Requests that break
* these rules are rejected with a 400 before any provider is called. `model` is any decision model in the proxy model list. OpenAI decision models accept this format
* too, LiteLLM translates it and keeps the answers keyed by question name. Streaming is not supported.
*
* [Docs](https://docs.litellm.ai/docs/decisions)
*/
post: operations["systemone_v1_systemone_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/threads": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Create Threads
* @description Create a thread.
*
* API Reference - https://platform.openai.com/docs/api-reference/threads/createThread
*/
post: operations["create_threads_v1_threads_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/threads/{thread_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Thread
* @description Retrieves a thread.
*
* API Reference - https://platform.openai.com/docs/api-reference/threads/getThread
*/
get: operations["get_thread_v1_threads__thread_id__get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/threads/{thread_id}/messages": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Messages
* @description Returns a list of messages for a given thread.
*
* API Reference - https://platform.openai.com/docs/api-reference/messages/listMessages
*/
get: operations["get_messages_v1_threads__thread_id__messages_get"];
put?: never;
/**
* Add Messages
* @description Create a message.
*
* API Reference - https://platform.openai.com/docs/api-reference/messages/createMessage
*/
post: operations["add_messages_v1_threads__thread_id__messages_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/threads/{thread_id}/runs": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Run Thread
* @description Create a run.
*
* API Reference: https://platform.openai.com/docs/api-reference/runs/createRun
*/
post: operations["run_thread_v1_threads__thread_id__runs_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/tool/list": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List Tools
* @description List all auto-discovered tools and their policies.
*
* Parameters:
* - input_policy: Optional filter — one of "trusted", "untrusted", "blocked"
*/
get: operations["list_tools_v1_tool_list_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/tool/policy": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Update Tool Policy
* @description Set the input_policy and/or output_policy for a tool (global), or block for a specific team/key (override).
*
* Parameters:
* - tool_name: str - The tool to update
* - input_policy: optional - "trusted" | "untrusted" | "blocked"
* - output_policy: optional - "trusted" | "untrusted"
* - team_id: optional - if set, create/update override for this team only
* - key_hash: optional - if set, create/update override for this key only
*/
post: operations["update_tool_policy_v1_tool_policy_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/tool/policy/options": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Tool Policy Options
* @description Return the available input and output policy options with descriptions.
* Static data — no DB call.
*/
get: operations["get_tool_policy_options_v1_tool_policy_options_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/tool/spend": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Tool Spend
* @description Spend attributed to each tool over a date range, for the Cost Optimization dashboard.
*
* Reads the ``LiteLLM_DailyToolSpend`` rollup, written at request time from invoked
* tools only (MCP tool calls and response tool_calls; declaring a tool without
* invoking it does not count). A request that invoked multiple tools counts its
* full spend toward each of them, so per-tool numbers are attributions and do not
* sum to a deduplicated total.
*
* ``by_tool`` is the top ``TOOL_SPEND_TOP_TOOLS`` tools by spend, aggregated in
* SQL, and ``daily`` covers only those tools, so the response is bounded by
* days x TOOL_SPEND_TOP_TOOLS regardless of the requested range or how many
* distinct tool names exist.
*/
get: operations["get_tool_spend_v1_tool_spend_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/tool/{tool_name}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Tool
* @description Get details for a single tool.
*/
get: operations["get_tool_v1_tool__tool_name__get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/tool/{tool_name}/detail": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Tool Detail
* @description Get a single tool with its policy overrides (for UI detail view).
*/
get: operations["get_tool_detail_v1_tool__tool_name__detail_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/tool/{tool_name}/logs": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Tool Usage Logs
* @description Return paginated spend logs for requests that invoked this tool (from SpendLogToolIndex).
* Declaring a tool in a request body without the model invoking it does not create an entry.
*/
get: operations["get_tool_usage_logs_v1_tool__tool_name__logs_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/tool/{tool_name}/overrides": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
post?: never;
/**
* Delete Tool Policy Override
* @description Remove a policy override for a tool. Specify the override by team_id or key_hash
* (exactly one required).
*/
delete: operations["delete_tool_policy_override_v1_tool__tool_name__overrides_delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/traces": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** List Agent Traces */
get: operations["list_agent_traces_v1_traces_get"];
put?: never;
/** Ingest Otlp Traces */
post: operations["ingest_otlp_traces_v1_traces_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/traces/agents": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** List Trace Agents */
get: operations["list_trace_agents_v1_traces_agents_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/traces/query": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/** Query Agent Traces */
post: operations["query_agent_traces_v1_traces_query_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/traces/query/help": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Help Agent Trace Queries */
get: operations["help_agent_trace_queries_v1_traces_query_help_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/traces/{trace_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get Agent Trace */
get: operations["get_agent_trace_v1_traces__trace_id__get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/traces/{trace_id}/spans/{span_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get Agent Trace Span */
get: operations["get_agent_trace_span_v1_traces__trace_id__spans__span_id__get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/traces/{trace_id}/spans/{span_id}/error": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get Agent Trace Span Error */
get: operations["get_agent_trace_span_error_v1_traces__trace_id__spans__span_id__error_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/unified_access_group": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** List Access Groups */
get: operations["list_access_groups_v1_unified_access_group_get"];
put?: never;
/** Create Access Group */
post: operations["create_access_group_v1_unified_access_group_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/unified_access_group/{access_group_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Get Access Group */
get: operations["get_access_group_v1_unified_access_group__access_group_id__get"];
/** Update Access Group */
put: operations["update_access_group_v1_unified_access_group__access_group_id__put"];
post?: never;
/** Delete Access Group */
delete: operations["delete_access_group_v1_unified_access_group__access_group_id__delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/vector_store/list": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List Vector Stores
* @description List all available vector stores with optional filtering and pagination.
* Combines both in-memory vector stores and those stored in the database.
* Database is the source of truth for stores it owns: deleted stores are removed from memory, updated stores
* sync to memory. Stores declared in the config file are owned by the config file, are always listed, and are
* never overwritten by database rows.
*
* Parameters:
* - page: int - Page number for pagination (default: 1)
* - page_size: int - Number of items per page (default: 100)
*/
get: operations["list_vector_stores_v1_vector_store_list_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/vector_stores": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Vector Store List
* @description List vector stores.
*
* API Reference:
* https://platform.openai.com/docs/api-reference/vector-stores/list
*/
get: operations["vector_store_list_v1_vector_stores_get"];
put?: never;
/**
* Vector Store Create
* @description Create a vector store.
*
* API Reference:
* https://platform.openai.com/docs/api-reference/vector-stores/create
*
* Supports target_model_names parameter for creating vector stores across multiple models:
* ```json
* {
* "name": "my-vector-store",
* "target_model_names": "gpt-4,gemini-2.0"
* }
* ```
*/
post: operations["vector_store_create_v1_vector_stores_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/vector_stores/{vector_store_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Vector Store Retrieve
* @description Retrieve a vector store.
*
* API Reference:
* https://platform.openai.com/docs/api-reference/vector-stores/retrieve
*/
get: operations["vector_store_retrieve_v1_vector_stores__vector_store_id__get"];
put?: never;
/**
* Vector Store Update
* @description Update a vector store.
*
* API Reference:
* https://platform.openai.com/docs/api-reference/vector-stores/modify
*/
post: operations["vector_store_update_v1_vector_stores__vector_store_id__post"];
/**
* Vector Store Delete
* @description Delete a vector store.
*
* API Reference:
* https://platform.openai.com/docs/api-reference/vector-stores/delete
*/
delete: operations["vector_store_delete_v1_vector_stores__vector_store_id__delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/vector_stores/{vector_store_id}/files": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Vector Store File List */
get: operations["vector_store_file_list_v1_vector_stores__vector_store_id__files_get"];
put?: never;
/** Vector Store File Create */
post: operations["vector_store_file_create_v1_vector_stores__vector_store_id__files_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/vector_stores/{vector_store_id}/files/{file_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Vector Store File Retrieve */
get: operations["vector_store_file_retrieve_v1_vector_stores__vector_store_id__files__file_id__get"];
put?: never;
/** Vector Store File Update */
post: operations["vector_store_file_update_v1_vector_stores__vector_store_id__files__file_id__post"];
/** Vector Store File Delete */
delete: operations["vector_store_file_delete_v1_vector_stores__vector_store_id__files__file_id__delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/vector_stores/{vector_store_id}/files/{file_id}/content": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Vector Store File Content */
get: operations["vector_store_file_content_v1_vector_stores__vector_store_id__files__file_id__content_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/vector_stores/{vector_store_id}/search": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Vector Store Search
* @description Search a vector store.
*
* API Reference:
* https://platform.openai.com/docs/api-reference/vector-stores/search
*/
post: operations["vector_store_search_v1_vector_stores__vector_store_id__search_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/videos": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Video List
* @description Video list endpoint for retrieving a list of videos.
*
* Follows the OpenAI Videos API spec:
* https://platform.openai.com/docs/api-reference/videos
*
* Example:
* ```bash
* curl -X GET "http://localhost:4000/v1/videos" -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*/
get: operations["video_list_v1_videos_get"];
put?: never;
/**
* Video Generation
* @description Video generation endpoint for creating videos from text prompts.
*
* Follows the OpenAI Videos API spec:
* https://platform.openai.com/docs/api-reference/videos
*
* Example:
* ```bash
* curl -X POST "http://localhost:4000/v1/videos" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -H "Content-Type: application/json" -d '{
* "model": "sora-2",
* "prompt": "A beautiful sunset over the ocean"
* }'
* ```
*/
post: operations["video_generation_v1_videos_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/videos/characters": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Video Create Character
* @description Create a character from an uploaded video file.
*
* Follows the OpenAI Videos API spec:
* https://platform.openai.com/docs/api-reference/videos/create-character
*
* Example:
* ```bash
* curl -X POST "http://localhost:4000/v1/videos/characters" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -F "video=@character_video.mp4" -F "name=my_character"
* ```
*/
post: operations["video_create_character_v1_videos_characters_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/videos/characters/{character_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Video Get Character
* @description Retrieve a character by ID.
*
* Follows the OpenAI Videos API spec:
* https://platform.openai.com/docs/api-reference/videos/get-character
*
* Example:
* ```bash
* curl -X GET "http://localhost:4000/v1/videos/characters/char_123" -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*/
get: operations["video_get_character_v1_videos_characters__character_id__get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/videos/edits": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Video Edit
* @description Create a video edit job.
*
* Follows the OpenAI Videos API spec:
* https://platform.openai.com/docs/api-reference/videos/create-edit
*
* Example:
* ```bash
* curl -X POST "http://localhost:4000/v1/videos/edits" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -H "Content-Type: application/json" -d '{"prompt": "Make it brighter", "video": {"id": "video_123"}}'
* ```
*/
post: operations["video_edit_v1_videos_edits_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/videos/extensions": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Video Extension
* @description Create a video extension.
*
* Follows the OpenAI Videos API spec:
* https://platform.openai.com/docs/api-reference/videos/create-extension
*
* Example:
* ```bash
* curl -X POST "http://localhost:4000/v1/videos/extensions" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -H "Content-Type: application/json" -d '{"prompt": "Continue the scene", "seconds": "5", "video": {"id": "video_123"}}'
* ```
*/
post: operations["video_extension_v1_videos_extensions_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/videos/{video_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Video Status
* @description Video status endpoint for retrieving video status and metadata.
*
* Follows the OpenAI Videos API spec:
* https://platform.openai.com/docs/api-reference/videos
*
* Example:
* ```bash
* curl -X GET "http://localhost:4000/v1/videos/video_123" -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*/
get: operations["video_status_v1_videos__video_id__get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/videos/{video_id}/content": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Video Content
* @description Video content endpoint for downloading video content.
*
* Follows the OpenAI Videos API spec:
* https://platform.openai.com/docs/api-reference/videos
*
* Example:
* ```bash
* curl -X GET "http://localhost:4000/v1/videos/{video_id}/content" -H "Authorization: Bearer $LITELLM_MASTER_KEY" --output video.mp4
* ```
*/
get: operations["video_content_v1_videos__video_id__content_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/videos/{video_id}/remix": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Video Remix
* @description Video remix endpoint for remixing existing videos with new prompts.
*
* Follows the OpenAI Videos API spec:
* https://platform.openai.com/docs/api-reference/videos
*
* Example:
* ```bash
* curl -X POST "http://localhost:4000/v1/videos/video_123/remix" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -H "Content-Type: application/json" -d '{
* "prompt": "A new version with different colors"
* }'
* ```
*/
post: operations["video_remix_v1_videos__video_id__remix_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/workflows/runs": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List Workflow Runs
* @description List workflow runs. Filter by workflow_type and/or status.
*
* Non-admin callers only see runs created by their own API key.
*/
get: operations["list_workflow_runs_v1_workflows_runs_get"];
put?: never;
/**
* Create Workflow Run
* @description Create a new workflow run. Returns run_id and session_id.
*
* The caller's API key token is stored as created_by so that non-admin keys
* can only see and modify their own runs.
*/
post: operations["create_workflow_run_v1_workflows_runs_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/workflows/runs/{run_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Workflow Run
* @description Get a workflow run with its most recent event.
*/
get: operations["get_workflow_run_v1_workflows_runs__run_id__get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
/**
* Update Workflow Run
* @description Update status, metadata, or output on a workflow run.
*/
patch: operations["update_workflow_run_v1_workflows_runs__run_id__patch"];
trace?: never;
};
"/v1/workflows/runs/{run_id}/events": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List Workflow Events
* @description Fetch event log for a run, ordered by sequence_number. Default limit 100, max 500.
*/
get: operations["list_workflow_events_v1_workflows_runs__run_id__events_get"];
put?: never;
/**
* Append Workflow Event
* @description Append an event to the run's event log. Also updates run.status if event_type maps to a status.
*
* Sequence numbers use optimistic concurrency: on a unique-constraint collision
* (concurrent append), retries up to _MAX_SEQUENCE_RETRIES times with a fresh MAX+1.
* The event+status update is atomic in a single DB transaction.
*/
post: operations["append_workflow_event_v1_workflows_runs__run_id__events_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1/workflows/runs/{run_id}/messages": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List Workflow Messages
* @description Fetch conversation history for a run, ordered by sequence_number. Default limit 100, max 500.
*/
get: operations["list_workflow_messages_v1_workflows_runs__run_id__messages_get"];
put?: never;
/**
* Append Workflow Message
* @description Append a conversation message. Stores full content (not truncated).
*
* Uses optimistic concurrency for sequence numbers.
*/
post: operations["append_workflow_message_v1_workflows_runs__run_id__messages_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1beta/agents": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List Gemini Agents
* @description List all custom agents on the Gemini side.
*
* Pass per-request Gemini credentials via the JSON-encoded
* ``litellm_params_template`` query parameter. Flat query parameters
* (e.g. ``?api_key=AIza...``) are intentionally ignored — see
* ``_merge_query_params_into_data`` for the rationale.
*
* ```bash
* curl "http://localhost:4000/v1beta/agents?litellm_params_template=%7B%22api_key%22%3A%22AIza...%22%7D" \
* -H "Authorization: Bearer sk-..."
* ```
*/
get: operations["list_gemini_agents_v1beta_agents_get"];
put?: never;
/**
* Create Gemini Agent
* @description Create a named custom agent on the Gemini side.
*
* Example:
* ```bash
* curl -X POST "http://localhost:4000/v1beta/agents" \
* -H "Authorization: Bearer sk-..." \
* -H "Content-Type: application/json" \
* -d '{
* "name": "my-custom-slides-agent",
* "base_agent": "waverunner",
* "instructions": "You are a helpful assistant that creates slides.",
* "base_environment": {
* "type": "remote",
* "sources": [
* {"type": "gcs", "source": "gs://eap-templates/slides-skill",
* "target": "/.agents/skills/slides-skill"}
* ]
* }
* }'
* ```
*/
post: operations["create_gemini_agent_v1beta_agents_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1beta/agents/{name}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Gemini Agent
* @description Get a specific custom agent by name.
*
* Pass per-request Gemini credentials via the JSON-encoded
* ``litellm_params_template`` query parameter. Flat query parameters
* (e.g. ``?api_key=AIza...``) are intentionally ignored — see
* ``_merge_query_params_into_data`` for the rationale.
*
* ```bash
* curl "http://localhost:4000/v1beta/agents/my-custom-slides-agent?litellm_params_template=%7B%22api_key%22%3A%22AIza...%22%7D" \
* -H "Authorization: Bearer sk-..."
* ```
*/
get: operations["get_gemini_agent_v1beta_agents__name__get"];
put?: never;
post?: never;
/**
* Delete Gemini Agent
* @description Delete a custom agent by name.
*
* Pass per-request Gemini credentials via the JSON-encoded
* ``litellm_params_template`` query parameter. Flat query parameters
* (e.g. ``?api_key=AIza...``) are intentionally ignored — see
* ``_merge_query_params_into_data`` for the rationale.
*
* ```bash
* curl -X DELETE "http://localhost:4000/v1beta/agents/my-custom-slides-agent?litellm_params_template=%7B%22api_key%22%3A%22AIza...%22%7D" \
* -H "Authorization: Bearer sk-..."
* ```
*/
delete: operations["delete_gemini_agent_v1beta_agents__name__delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1beta/agents/{name}/versions": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List Gemini Agent Versions
* @description List versions of a custom agent.
*
* Pass per-request Gemini credentials via the JSON-encoded
* ``litellm_params_template`` query parameter. Flat query parameters
* (e.g. ``?api_key=AIza...``) are intentionally ignored — see
* ``_merge_query_params_into_data`` for the rationale.
*
* ```bash
* curl "http://localhost:4000/v1beta/agents/my-custom-slides-agent/versions?litellm_params_template=%7B%22api_key%22%3A%22AIza...%22%7D" \
* -H "Authorization: Bearer sk-..."
* ```
*/
get: operations["list_gemini_agent_versions_v1beta_agents__name__versions_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1beta/interactions": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Create Interaction
* @description Create a new interaction using Google's Interactions API.
*
* Per OpenAPI spec: POST /{api_version}/interactions
*
* Supports both model interactions and agent interactions:
* - Model: Provide `model` parameter (e.g., "gemini-2.5-flash")
* - Agent: Provide `agent` parameter (e.g., "deep-research-pro-preview-12-2025")
*
* Example:
* ```bash
* curl -X POST "http://localhost:4000/v1beta/interactions" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -H "Content-Type: application/json" -d '{
* "model": "gemini/gemini-2.5-flash",
* "input": "Hello, how are you?"
* }'
* ```
*/
post: operations["create_interaction_v1beta_interactions_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1beta/interactions/{interaction_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Interaction
* @description Get an interaction by ID.
*
* Per OpenAPI spec: GET /{api_version}/interactions/{interaction_id}
*/
get: operations["get_interaction_v1beta_interactions__interaction_id__get"];
put?: never;
post?: never;
/**
* Delete Interaction
* @description Delete an interaction by ID.
*
* Per OpenAPI spec: DELETE /{api_version}/interactions/{interaction_id}
*/
delete: operations["delete_interaction_v1beta_interactions__interaction_id__delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1beta/interactions/{interaction_id}/cancel": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Cancel Interaction
* @description Cancel an interaction by ID.
*
* Per OpenAPI spec: POST /{api_version}/interactions/{interaction_id}/cancel
*/
post: operations["cancel_interaction_v1beta_interactions__interaction_id__cancel_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1beta/models/{model_name}:countTokens": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Google Count Tokens
* @description ```json
* return {
* "totalTokens": 31,
* "totalBillableCharacters": 96,
* "promptTokensDetails": [
* {
* "modality": "TEXT",
* "tokenCount": 31
* }
* ]
* }
* ```
*/
post: operations["google_count_tokens_v1beta_models__model_name__countTokens_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1beta/models/{model_name}:generateContent": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/** Google Generate Content */
post: operations["google_generate_content_v1beta_models__model_name__generateContent_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v1beta/models/{model_name}:streamGenerateContent": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/** Google Stream Generate Content */
post: operations["google_stream_generate_content_v1beta_models__model_name__streamGenerateContent_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v2/guardrails/list": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List Guardrails V2
* @description List the guardrails that are available in the database using GuardrailRegistry
*
* 👉 [Guardrail docs](https://docs.litellm.ai/docs/proxy/guardrails/quick_start)
*
* Example Request:
* ```bash
* curl -X GET "http://localhost:4000/v2/guardrails/list" -H "Authorization: Bearer <your_api_key>"
* ```
*
* Example Response:
* ```json
* {
* "guardrails": [
* {
* "guardrail_id": "123e4567-e89b-12d3-a456-426614174000",
* "guardrail_name": "my-bedrock-guard",
* "litellm_params": {
* "guardrail": "bedrock",
* "mode": "pre_call",
* "guardrailIdentifier": "ff6ujrregl1q",
* "guardrailVersion": "DRAFT",
* "default_on": true
* },
* "guardrail_info": {
* "description": "Bedrock content moderation guardrail"
* }
* }
* ]
* }
* ```
*/
get: operations["list_guardrails_v2_v2_guardrails_list_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v2/key/info": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Info Key Fn V2
* @description Retrieve information about a list of keys.
*
* **New endpoint**. Currently admin only.
* Parameters:
* keys: Optional[list] = body parameter representing the key(s) in the request
* user_api_key_dict: UserAPIKeyAuth = Dependency representing the user's API key
* Returns:
* Dict containing the key and its associated information
*
* Example Curl:
* ```
* curl -X GET "http://0.0.0.0:4000/key/info" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -d {"keys": ["sk-1", "sk-2", "sk-3"]}
* ```
*/
post: operations["info_key_fn_v2_v2_key_info_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v2/login": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/** Login V2 */
post: operations["login_v2_v2_login_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v2/model/info": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Model Info V2
* @description Paginated model metadata for proxy deployments (pricing, provider, team access).
*
* Returns configured router deployments with enriched `model_info` (costs, provider,
* context window, etc.). Sensitive fields such as API keys and api_base are omitted.
*
* Query parameters:
* model: Filter to a single public `model_name`.
* user_models_only: When true, only return models created by the calling user.
* include_team_models: When true, populate `access_via_team_ids` and `direct_access`
* on each model and filter to deployments the caller can use.
* page / size: Pagination controls (defaults: page=1, size=50).
* search: Case-insensitive partial match on model name or team public name.
* modelId: Return a single deployment by LiteLLM model id.
* teamId: Filter to models with direct access or team membership for this team id.
* sortBy / sortOrder: Sort by model_name, created_at, updated_at, costs, or status.
* blocked: Filter by routing status (false = active, true = paused).
* access_group: Only return deployments in this model access group.
* wildcard_only: Only return deployments whose `model_name` contains `*`.
*
* Example request:
* ```
* curl -X GET 'http://localhost:4000/v2/model/info?include_team_models=true&page=1&size=50' \
* --header "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*
* Example response:
* ```json
* {
* "data": [
* {
* "model_name": "gpt-4",
* "litellm_params": {"model": "openai/gpt-4.1"},
* "model_info": {
* "id": "abc123",
* "litellm_provider": "openai",
* "access_via_team_ids": ["team-1"],
* "direct_access": true
* }
* }
* ],
* "total_count": 1,
* "current_page": 1,
* "total_pages": 1,
* "size": 50
* }
* ```
*/
get: operations["model_info_v2_v2_model_info_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v2/organization/{organization_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
/**
* Update Organization V2
* @description Partial update of an organization (RESTful PATCH, RFC 7396 merge-patch semantics).
*
* A sent field is written and an omitted one is left untouched (presence is read from
* ``model_fields_set``). Clear tokens are per field: budget limits and ``metadata`` clear with
* ``null``, ``models`` with ``[]``, and ``object_permission`` with ``null`` (it merges when sent,
* so an empty ``{}`` is rejected). ``organization_alias`` is required and cannot be cleared.
* Validation failures return 422; the object-permission upsert, budget-row write, and
* org-row write are one transaction.
*/
patch: operations["update_organization_v2_v2_organization__organization_id__patch"];
trace?: never;
};
"/v2/rerank": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/** Rerank */
post: operations["rerank_v2_rerank_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v2/team/list": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List Team V2
* @description Get a paginated list of teams with filtering and sorting options.
*
* Parameters:
* user_id: Optional[str]
* Only return teams which this user belongs to
* organization_id: Optional[str]
* Only return teams which belong to this organization
* team_id: Optional[str]
* Filter teams by exact team_id match
* team_alias: Optional[str]
* Filter teams by partial team_alias match
* page: int
* The page number to return
* page_size: int
* The number of items per page
* sort_by: Optional[str]
* Column to sort by (e.g. 'team_id', 'team_alias', 'created_at')
* sort_order: str
* Sort order ('asc' or 'desc')
* status: Optional[str]
* Filter by status. Currently supports "deleted" to query deleted teams.
*/
get: operations["list_team_v2_v2_team_list_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v2/user/info": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* User Info V2
* @description Lightweight endpoint to get user info. Returns only the user object — no keys, no teams objects.
*
* This is the v2 replacement for /user/info, designed to avoid the "god endpoint" problem
* where the old endpoint loaded all keys and teams into memory.
*
* Note on `spend`: this is the user's running budget counter, which the budget reset job
* resets whenever `budget_reset_at` elapses (see `budget_duration`): to zero by default,
* or to the overage above `max_budget` when `budget_rollover` is enabled. It is NOT
* lifetime or per-period historical spend. For historical spend over a date range, use
* `/user/daily/activity` or `/user/daily/activity/aggregated`, which read daily spend
* records that only ever accumulate and are never reset. The two values are expected to
* diverge once a budget reset has occurred within the queried period.
*
* Access control:
* - Proxy admins can query any user
* - Team admins can query users within their teams
* - Internal users can only query themselves (omit user_id or pass own)
* - Returns 404 for non-existent users or unauthorized access
*
* Example request:
* ```
* curl -X GET 'http://localhost:4000/v2/user/info?user_id=user123' \
* --header "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*/
get: operations["user_info_v2_v2_user_info_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v3/login": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/** Login V3 */
post: operations["login_v3_v3_login_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/v3/login/exchange": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/** Login V3 Exchange */
post: operations["login_v3_exchange_v3_login_exchange_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/vantage/delete": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
post?: never;
/**
* Delete Vantage Settings
* @description Delete Vantage settings from the database.
*
* Only admin users can delete Vantage settings.
*/
delete: operations["delete_vantage_settings_vantage_delete_delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/vantage/dry-run": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Vantage Dry Run Export
* @description Perform a dry run export using the Vantage logger.
*
* Returns the data that would be exported without actually sending it to Vantage.
*
* Parameters:
* - limit: Limit on number of records to preview (default: 500)
*
* Only admin users can perform Vantage exports.
*/
post: operations["vantage_dry_run_export_vantage_dry_run_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/vantage/export": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Vantage Export
* @description Perform an actual export using the Vantage logger.
*
* Exports usage data in FOCUS CSV format to the Vantage API.
*
* Parameters:
* - limit: Optional limit on number of records to export
* - start_time_utc: Optional start time for data export
* - end_time_utc: Optional end time for data export
*
* Only admin users can perform Vantage exports.
*/
post: operations["vantage_export_vantage_export_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/vantage/init": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Init Vantage Settings
* @description Initialize Vantage settings and store in the database.
*
* Parameters:
* - api_key: Vantage API key for authentication
* - integration_token: Vantage integration token for the cost-import endpoint
* - base_url: Vantage API base URL (default: https://api.vantage.sh)
*
* Only admin users can configure Vantage settings.
*/
post: operations["init_vantage_settings_vantage_init_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/vantage/settings": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get Vantage Settings
* @description View current Vantage settings.
*
* Returns the current Vantage configuration with the API key masked for security.
* Only admin users (Proxy Admin or Admin Viewer) can view Vantage settings.
*/
get: operations["get_vantage_settings_vantage_settings_get"];
/**
* Update Vantage Settings
* @description Update existing Vantage settings.
*
* Allows updating individual Vantage configuration fields without requiring all fields.
* Only admin users can update Vantage settings.
*/
put: operations["update_vantage_settings_vantage_settings_put"];
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/vector_store/delete": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Delete Vector Store
* @description Delete a vector store from both database and in-memory registry.
*
* Parameters:
* - vector_store_id: str - ID of the vector store to delete
*/
post: operations["delete_vector_store_vector_store_delete_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/vector_store/info": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Get Vector Store Info
* @description Return a single vector store's details
*/
post: operations["get_vector_store_info_vector_store_info_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/vector_store/list": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List Vector Stores
* @description List all available vector stores with optional filtering and pagination.
* Combines both in-memory vector stores and those stored in the database.
* Database is the source of truth for stores it owns: deleted stores are removed from memory, updated stores
* sync to memory. Stores declared in the config file are owned by the config file, are always listed, and are
* never overwritten by database rows.
*
* Parameters:
* - page: int - Page number for pagination (default: 1)
* - page_size: int - Number of items per page (default: 100)
*/
get: operations["list_vector_stores_vector_store_list_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/vector_store/new": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* New Vector Store
* @description Create a new vector store.
*
* Parameters:
* - vector_store_id: str - Unique identifier for the vector store
* - custom_llm_provider: str - Provider of the vector store
* - vector_store_name: Optional[str] - Name of the vector store
* - vector_store_description: Optional[str] - Description of the vector store
* - vector_store_metadata: Optional[Dict] - Additional metadata for the vector store
*/
post: operations["new_vector_store_vector_store_new_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/vector_store/update": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Update Vector Store
* @description Update vector store details in both database and in-memory registry.
* The updated data is immediately synchronized to the in-memory registry.
*/
post: operations["update_vector_store_vector_store_update_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/vector_stores": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Vector Store List
* @description List vector stores.
*
* API Reference:
* https://platform.openai.com/docs/api-reference/vector-stores/list
*/
get: operations["vector_store_list_vector_stores_get"];
put?: never;
/**
* Vector Store Create
* @description Create a vector store.
*
* API Reference:
* https://platform.openai.com/docs/api-reference/vector-stores/create
*
* Supports target_model_names parameter for creating vector stores across multiple models:
* ```json
* {
* "name": "my-vector-store",
* "target_model_names": "gpt-4,gemini-2.0"
* }
* ```
*/
post: operations["vector_store_create_vector_stores_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/vector_stores/{vector_store_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Vector Store Retrieve
* @description Retrieve a vector store.
*
* API Reference:
* https://platform.openai.com/docs/api-reference/vector-stores/retrieve
*/
get: operations["vector_store_retrieve_vector_stores__vector_store_id__get"];
put?: never;
/**
* Vector Store Update
* @description Update a vector store.
*
* API Reference:
* https://platform.openai.com/docs/api-reference/vector-stores/modify
*/
post: operations["vector_store_update_vector_stores__vector_store_id__post"];
/**
* Vector Store Delete
* @description Delete a vector store.
*
* API Reference:
* https://platform.openai.com/docs/api-reference/vector-stores/delete
*/
delete: operations["vector_store_delete_vector_stores__vector_store_id__delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/vector_stores/{vector_store_id}/files": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Vector Store File List */
get: operations["vector_store_file_list_vector_stores__vector_store_id__files_get"];
put?: never;
/** Vector Store File Create */
post: operations["vector_store_file_create_vector_stores__vector_store_id__files_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/vector_stores/{vector_store_id}/files/{file_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Vector Store File Retrieve */
get: operations["vector_store_file_retrieve_vector_stores__vector_store_id__files__file_id__get"];
put?: never;
/** Vector Store File Update */
post: operations["vector_store_file_update_vector_stores__vector_store_id__files__file_id__post"];
/** Vector Store File Delete */
delete: operations["vector_store_file_delete_vector_stores__vector_store_id__files__file_id__delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/vector_stores/{vector_store_id}/files/{file_id}/content": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Vector Store File Content */
get: operations["vector_store_file_content_vector_stores__vector_store_id__files__file_id__content_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/vector_stores/{vector_store_id}/search": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Vector Store Search
* @description Search a vector store.
*
* API Reference:
* https://platform.openai.com/docs/api-reference/vector-stores/search
*/
post: operations["vector_store_search_vector_stores__vector_store_id__search_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/vertex_ai/discovery/{endpoint}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Vertex Discovery Proxy Route
* @description Call any vertex discovery endpoint using the proxy.
*
* Just use `{PROXY_BASE_URL}/vertex_ai/discovery/{endpoint:path}`
*
* Target url: `https://discoveryengine.googleapis.com`
*/
get: operations["vertex_discovery_proxy_route_vertex_ai_discovery__endpoint__get"];
/**
* Vertex Discovery Proxy Route
* @description Call any vertex discovery endpoint using the proxy.
*
* Just use `{PROXY_BASE_URL}/vertex_ai/discovery/{endpoint:path}`
*
* Target url: `https://discoveryengine.googleapis.com`
*/
put: operations["vertex_discovery_proxy_route_vertex_ai_discovery__endpoint__put"];
/**
* Vertex Discovery Proxy Route
* @description Call any vertex discovery endpoint using the proxy.
*
* Just use `{PROXY_BASE_URL}/vertex_ai/discovery/{endpoint:path}`
*
* Target url: `https://discoveryengine.googleapis.com`
*/
post: operations["vertex_discovery_proxy_route_vertex_ai_discovery__endpoint__post"];
/**
* Vertex Discovery Proxy Route
* @description Call any vertex discovery endpoint using the proxy.
*
* Just use `{PROXY_BASE_URL}/vertex_ai/discovery/{endpoint:path}`
*
* Target url: `https://discoveryengine.googleapis.com`
*/
delete: operations["vertex_discovery_proxy_route_vertex_ai_discovery__endpoint__delete"];
options?: never;
head?: never;
/**
* Vertex Discovery Proxy Route
* @description Call any vertex discovery endpoint using the proxy.
*
* Just use `{PROXY_BASE_URL}/vertex_ai/discovery/{endpoint:path}`
*
* Target url: `https://discoveryengine.googleapis.com`
*/
patch: operations["vertex_discovery_proxy_route_vertex_ai_discovery__endpoint__patch"];
trace?: never;
};
"/vertex_ai/live": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* WebSocket: vertex_ai_live_passthrough_endpoint
* @description WebSocket connection endpoint
*/
get: operations["websocket_vertex_ai_live_passthrough_endpoint"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/vertex_ai/{endpoint}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Vertex Proxy Route
* @description Call LiteLLM proxy via Vertex AI SDK.
*
* [Docs](https://docs.litellm.ai/docs/pass_through/vertex_ai)
*/
get: operations["vertex_proxy_route_vertex_ai__endpoint__get"];
/**
* Vertex Proxy Route
* @description Call LiteLLM proxy via Vertex AI SDK.
*
* [Docs](https://docs.litellm.ai/docs/pass_through/vertex_ai)
*/
put: operations["vertex_proxy_route_vertex_ai__endpoint__put"];
/**
* Vertex Proxy Route
* @description Call LiteLLM proxy via Vertex AI SDK.
*
* [Docs](https://docs.litellm.ai/docs/pass_through/vertex_ai)
*/
post: operations["vertex_proxy_route_vertex_ai__endpoint__post"];
/**
* Vertex Proxy Route
* @description Call LiteLLM proxy via Vertex AI SDK.
*
* [Docs](https://docs.litellm.ai/docs/pass_through/vertex_ai)
*/
delete: operations["vertex_proxy_route_vertex_ai__endpoint__delete"];
options?: never;
head?: never;
/**
* Vertex Proxy Route
* @description Call LiteLLM proxy via Vertex AI SDK.
*
* [Docs](https://docs.litellm.ai/docs/pass_through/vertex_ai)
*/
patch: operations["vertex_proxy_route_vertex_ai__endpoint__patch"];
trace?: never;
};
"/videos": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Video List
* @description Video list endpoint for retrieving a list of videos.
*
* Follows the OpenAI Videos API spec:
* https://platform.openai.com/docs/api-reference/videos
*
* Example:
* ```bash
* curl -X GET "http://localhost:4000/v1/videos" -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*/
get: operations["video_list_videos_get"];
put?: never;
/**
* Video Generation
* @description Video generation endpoint for creating videos from text prompts.
*
* Follows the OpenAI Videos API spec:
* https://platform.openai.com/docs/api-reference/videos
*
* Example:
* ```bash
* curl -X POST "http://localhost:4000/v1/videos" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -H "Content-Type: application/json" -d '{
* "model": "sora-2",
* "prompt": "A beautiful sunset over the ocean"
* }'
* ```
*/
post: operations["video_generation_videos_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/videos/characters": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Video Create Character
* @description Create a character from an uploaded video file.
*
* Follows the OpenAI Videos API spec:
* https://platform.openai.com/docs/api-reference/videos/create-character
*
* Example:
* ```bash
* curl -X POST "http://localhost:4000/v1/videos/characters" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -F "video=@character_video.mp4" -F "name=my_character"
* ```
*/
post: operations["video_create_character_videos_characters_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/videos/characters/{character_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Video Get Character
* @description Retrieve a character by ID.
*
* Follows the OpenAI Videos API spec:
* https://platform.openai.com/docs/api-reference/videos/get-character
*
* Example:
* ```bash
* curl -X GET "http://localhost:4000/v1/videos/characters/char_123" -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*/
get: operations["video_get_character_videos_characters__character_id__get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/videos/edits": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Video Edit
* @description Create a video edit job.
*
* Follows the OpenAI Videos API spec:
* https://platform.openai.com/docs/api-reference/videos/create-edit
*
* Example:
* ```bash
* curl -X POST "http://localhost:4000/v1/videos/edits" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -H "Content-Type: application/json" -d '{"prompt": "Make it brighter", "video": {"id": "video_123"}}'
* ```
*/
post: operations["video_edit_videos_edits_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/videos/extensions": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Video Extension
* @description Create a video extension.
*
* Follows the OpenAI Videos API spec:
* https://platform.openai.com/docs/api-reference/videos/create-extension
*
* Example:
* ```bash
* curl -X POST "http://localhost:4000/v1/videos/extensions" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -H "Content-Type: application/json" -d '{"prompt": "Continue the scene", "seconds": "5", "video": {"id": "video_123"}}'
* ```
*/
post: operations["video_extension_videos_extensions_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/videos/{video_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Video Status
* @description Video status endpoint for retrieving video status and metadata.
*
* Follows the OpenAI Videos API spec:
* https://platform.openai.com/docs/api-reference/videos
*
* Example:
* ```bash
* curl -X GET "http://localhost:4000/v1/videos/video_123" -H "Authorization: Bearer $LITELLM_MASTER_KEY"
* ```
*/
get: operations["video_status_videos__video_id__get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/videos/{video_id}/content": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Video Content
* @description Video content endpoint for downloading video content.
*
* Follows the OpenAI Videos API spec:
* https://platform.openai.com/docs/api-reference/videos
*
* Example:
* ```bash
* curl -X GET "http://localhost:4000/v1/videos/{video_id}/content" -H "Authorization: Bearer $LITELLM_MASTER_KEY" --output video.mp4
* ```
*/
get: operations["video_content_videos__video_id__content_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/videos/{video_id}/remix": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Video Remix
* @description Video remix endpoint for remixing existing videos with new prompts.
*
* Follows the OpenAI Videos API spec:
* https://platform.openai.com/docs/api-reference/videos
*
* Example:
* ```bash
* curl -X POST "http://localhost:4000/v1/videos/video_123/remix" -H "Authorization: Bearer $LITELLM_MASTER_KEY" -H "Content-Type: application/json" -d '{
* "prompt": "A new version with different colors"
* }'
* ```
*/
post: operations["video_remix_videos__video_id__remix_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/vllm/{endpoint}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Vllm Proxy Route
* @description [Docs](https://docs.litellm.ai/docs/pass_through/vllm)
*/
get: operations["vllm_proxy_route_vllm__endpoint__get"];
/**
* Vllm Proxy Route
* @description [Docs](https://docs.litellm.ai/docs/pass_through/vllm)
*/
put: operations["vllm_proxy_route_vllm__endpoint__put"];
/**
* Vllm Proxy Route
* @description [Docs](https://docs.litellm.ai/docs/pass_through/vllm)
*/
post: operations["vllm_proxy_route_vllm__endpoint__post"];
/**
* Vllm Proxy Route
* @description [Docs](https://docs.litellm.ai/docs/pass_through/vllm)
*/
delete: operations["vllm_proxy_route_vllm__endpoint__delete"];
options?: never;
head?: never;
/**
* Vllm Proxy Route
* @description [Docs](https://docs.litellm.ai/docs/pass_through/vllm)
*/
patch: operations["vllm_proxy_route_vllm__endpoint__patch"];
trace?: never;
};
"/watsonx/{endpoint}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Watsonx Proxy Route
* @description Watsonx pass-through endpoint.
* Allows using Watsonx APIs with automatic IAM token management and version parameter injection.
*
* Example:
* POST /watsonx/ml/v1/text/tokenization
* POST /watsonx/ml/v1/text/generation
*/
get: operations["watsonx_proxy_route_watsonx__endpoint__get"];
/**
* Watsonx Proxy Route
* @description Watsonx pass-through endpoint.
* Allows using Watsonx APIs with automatic IAM token management and version parameter injection.
*
* Example:
* POST /watsonx/ml/v1/text/tokenization
* POST /watsonx/ml/v1/text/generation
*/
put: operations["watsonx_proxy_route_watsonx__endpoint__put"];
/**
* Watsonx Proxy Route
* @description Watsonx pass-through endpoint.
* Allows using Watsonx APIs with automatic IAM token management and version parameter injection.
*
* Example:
* POST /watsonx/ml/v1/text/tokenization
* POST /watsonx/ml/v1/text/generation
*/
post: operations["watsonx_proxy_route_watsonx__endpoint__post"];
/**
* Watsonx Proxy Route
* @description Watsonx pass-through endpoint.
* Allows using Watsonx APIs with automatic IAM token management and version parameter injection.
*
* Example:
* POST /watsonx/ml/v1/text/tokenization
* POST /watsonx/ml/v1/text/generation
*/
delete: operations["watsonx_proxy_route_watsonx__endpoint__delete"];
options?: never;
head?: never;
/**
* Watsonx Proxy Route
* @description Watsonx pass-through endpoint.
* Allows using Watsonx APIs with automatic IAM token management and version parameter injection.
*
* Example:
* POST /watsonx/ml/v1/text/tokenization
* POST /watsonx/ml/v1/text/generation
*/
patch: operations["watsonx_proxy_route_watsonx__endpoint__patch"];
trace?: never;
};
"/{mcp_server_name}/authorize": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/** Authorize */
get: operations["authorize__mcp_server_name__authorize_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/{mcp_server_name}/mcp": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Dynamic Mcp Route
* @description Handle /{name}/mcp for MCP server aliases, toolsets, MCP access group tags, and comma-separated lists.
*
* Resolution order:
* 1. Registered MCP server alias / name
* 2. Comma-separated list
* 3. Toolset name (DB lookup, cached)
* 4. MCP access group tag (DB lookup, cached)
*/
get: operations["dynamic_mcp_route__mcp_server_name__mcp_get"];
/**
* Dynamic Mcp Route
* @description Handle /{name}/mcp for MCP server aliases, toolsets, MCP access group tags, and comma-separated lists.
*
* Resolution order:
* 1. Registered MCP server alias / name
* 2. Comma-separated list
* 3. Toolset name (DB lookup, cached)
* 4. MCP access group tag (DB lookup, cached)
*/
put: operations["dynamic_mcp_route__mcp_server_name__mcp_put"];
/**
* Dynamic Mcp Route
* @description Handle /{name}/mcp for MCP server aliases, toolsets, MCP access group tags, and comma-separated lists.
*
* Resolution order:
* 1. Registered MCP server alias / name
* 2. Comma-separated list
* 3. Toolset name (DB lookup, cached)
* 4. MCP access group tag (DB lookup, cached)
*/
post: operations["dynamic_mcp_route__mcp_server_name__mcp_post"];
/**
* Dynamic Mcp Route
* @description Handle /{name}/mcp for MCP server aliases, toolsets, MCP access group tags, and comma-separated lists.
*
* Resolution order:
* 1. Registered MCP server alias / name
* 2. Comma-separated list
* 3. Toolset name (DB lookup, cached)
* 4. MCP access group tag (DB lookup, cached)
*/
delete: operations["dynamic_mcp_route__mcp_server_name__mcp_delete"];
/**
* Dynamic Mcp Route
* @description Handle /{name}/mcp for MCP server aliases, toolsets, MCP access group tags, and comma-separated lists.
*
* Resolution order:
* 1. Registered MCP server alias / name
* 2. Comma-separated list
* 3. Toolset name (DB lookup, cached)
* 4. MCP access group tag (DB lookup, cached)
*/
options: operations["dynamic_mcp_route__mcp_server_name__mcp_options"];
/**
* Dynamic Mcp Route
* @description Handle /{name}/mcp for MCP server aliases, toolsets, MCP access group tags, and comma-separated lists.
*
* Resolution order:
* 1. Registered MCP server alias / name
* 2. Comma-separated list
* 3. Toolset name (DB lookup, cached)
* 4. MCP access group tag (DB lookup, cached)
*/
head: operations["dynamic_mcp_route__mcp_server_name__mcp_head"];
/**
* Dynamic Mcp Route
* @description Handle /{name}/mcp for MCP server aliases, toolsets, MCP access group tags, and comma-separated lists.
*
* Resolution order:
* 1. Registered MCP server alias / name
* 2. Comma-separated list
* 3. Toolset name (DB lookup, cached)
* 4. MCP access group tag (DB lookup, cached)
*/
patch: operations["dynamic_mcp_route__mcp_server_name__mcp_patch"];
trace?: never;
};
"/{mcp_server_name}/register": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/** Register Client */
post: operations["register_client__mcp_server_name__register_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/{mcp_server_name}/token": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Token Endpoint
* @description Accept the authorization code from client and exchange it for OAuth token.
* Supports PKCE flow by forwarding code_verifier to upstream provider.
*
* 1. Call the token endpoint with PKCE parameters
* 2. Store the user's token in the db - and generate a LiteLLM virtual key
* 3. Return the token
* 4. Return a virtual key in this response
*/
post: operations["token_endpoint__mcp_server_name__token_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/{provider}/v1/batches": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List Batches
* @description Lists
* This is the equivalent of GET https://api.openai.com/v1/batches/
* Supports Identical Params as: https://platform.openai.com/docs/api-reference/batch/list
*
* Example Curl
* ```
* curl http://localhost:4000/v1/batches?limit=2 -H "Authorization: Bearer $LITELLM_MASTER_KEY" -H "Content-Type: application/json"
* ```
*/
get: operations["list_batches__provider__v1_batches_get"];
put?: never;
/**
* Create Batch
* @description Create large batches of API requests for asynchronous processing.
* This is the equivalent of POST https://api.openai.com/v1/batch
* Supports Identical Params as: https://platform.openai.com/docs/api-reference/batch
*
* Example Curl
* ```
* curl http://localhost:4000/v1/batches -H "Authorization: Bearer $LITELLM_MASTER_KEY" -H "Content-Type: application/json" -d '{
* "input_file_id": "file-abc123",
* "endpoint": "/v1/chat/completions",
* "completion_window": "24h"
* }'
* ```
*/
post: operations["create_batch__provider__v1_batches_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/{provider}/v1/batches/{batch_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Retrieve Batch
* @description Retrieves a batch.
* This is the equivalent of GET https://api.openai.com/v1/batches/{batch_id}
* Supports Identical Params as: https://platform.openai.com/docs/api-reference/batch/retrieve
*
* Example Curl
* ```
* curl http://localhost:4000/v1/batches/batch_abc123 -H "Authorization: Bearer $LITELLM_MASTER_KEY" -H "Content-Type: application/json"
* ```
*/
get: operations["retrieve_batch__provider__v1_batches__batch_id__get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/{provider}/v1/batches/{batch_id}/cancel": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
get?: never;
put?: never;
/**
* Cancel Batch
* @description Cancel a batch.
* This is the equivalent of POST https://api.openai.com/v1/batches/{batch_id}/cancel
*
* Supports Identical Params as: https://platform.openai.com/docs/api-reference/batch/cancel
*
* Example Curl
* ```
* curl http://localhost:4000/v1/batches/batch_abc123/cancel -H "Authorization: Bearer $LITELLM_MASTER_KEY" -H "Content-Type: application/json" -X POST
*
* ```
*/
post: operations["cancel_batch__provider__v1_batches__batch_id__cancel_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/{provider}/v1/files": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* List Files
* @description Returns information about a specific file. that can be used across - Assistants API, Batch API
* This is the equivalent of GET https://api.openai.com/v1/files/
*
* Supports Identical Params as: https://platform.openai.com/docs/api-reference/files/list
*
* Example Curl
* ```
* curl http://localhost:4000/v1/files -H "Authorization: Bearer $LITELLM_MASTER_KEY"
*
* ```
*/
get: operations["list_files__provider__v1_files_get"];
put?: never;
/**
* Create File
* @description Upload a file that can be used across - Assistants API, Batch API
* This is the equivalent of POST https://api.openai.com/v1/files
*
* Supports Identical Params as: https://platform.openai.com/docs/api-reference/files/create
*
* Example Curl
* ```
* curl http://localhost:4000/v1/files -H "Authorization: Bearer $LITELLM_MASTER_KEY" -F purpose="batch" -F file="@mydata.jsonl"
* -F expires_after[anchor]="created_at" -F expires_after[seconds]=2592000
* ```
*/
post: operations["create_file__provider__v1_files_post"];
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/{provider}/v1/files/{file_id}": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get File
* @description Returns information about a specific file. that can be used across - Assistants API, Batch API
* This is the equivalent of GET https://api.openai.com/v1/files/{file_id}
*
* Supports Identical Params as: https://platform.openai.com/docs/api-reference/files/retrieve
*
* Example Curl
* ```
* curl http://localhost:4000/v1/files/file-abc123 -H "Authorization: Bearer $LITELLM_MASTER_KEY"
*
* ```
*/
get: operations["get_file__provider__v1_files__file_id__get"];
put?: never;
post?: never;
/**
* Delete File
* @description Deletes a specified file. that can be used across - Assistants API, Batch API
* This is the equivalent of DELETE https://api.openai.com/v1/files/{file_id}
*
* Supports Identical Params as: https://platform.openai.com/docs/api-reference/files/delete
*
* Example Curl
* ```
* curl http://localhost:4000/v1/files/file-abc123 -X DELETE -H "Authorization: Bearer $OPENAI_API_KEY"
*
* ```
*/
delete: operations["delete_file__provider__v1_files__file_id__delete"];
options?: never;
head?: never;
patch?: never;
trace?: never;
};
"/{provider}/v1/files/{file_id}/content": {
parameters: {
query?: never;
header?: never;
path?: never;
cookie?: never;
};
/**
* Get File Content
* @description Returns information about a specific file. that can be used across - Assistants API, Batch API
* This is the equivalent of GET https://api.openai.com/v1/files/{file_id}/content
*
* Supports Identical Params as: https://platform.openai.com/docs/api-reference/files/retrieve-contents
*
* Example Curl
* ```
* curl http://localhost:4000/v1/files/file-abc123/content -H "Authorization: Bearer $LITELLM_MASTER_KEY"
*
* ```
*/
get: operations["get_file_content__provider__v1_files__file_id__content_get"];
put?: never;
post?: never;
delete?: never;
options?: never;
head?: never;
patch?: never;
trace?: never;
};
}
export type webhooks = Record<string, never>;
export interface components {
schemas: {
/**
* APIKeySecurityScheme
* @description Defines a security scheme using an API key.
*/
APIKeySecurityScheme: {
/** Description */
description?: string | null;
/**
* In
* @enum {string}
*/
in_: "query" | "header" | "cookie";
/** Name */
name: string;
/**
* Type
* @constant
*/
type: "apiKey";
};
/**
* AccessControl_UI_AccessMode
* @description Model for Controlling UI Access Mode via SSO Groups
*/
AccessControl_UI_AccessMode: {
/** Restricted Sso Group */
restricted_sso_group: string;
/** Sso Group Jwt Field */
sso_group_jwt_field: string;
/**
* Type
* @constant
*/
type: "restricted_sso_group";
};
/** AccessGroupBudget */
AccessGroupBudget: {
/** Budget Duration */
budget_duration?: string | null;
/** Budget Id */
budget_id: string;
/** Budget Reset At */
budget_reset_at?: string | null;
/** Max Budget */
max_budget?: number | null;
/** Soft Budget */
soft_budget?: number | null;
};
/** AccessGroupBudgetRequest */
AccessGroupBudgetRequest: {
/** Budget Duration */
budget_duration?: string | null;
/** Budget Id */
budget_id?: string | null;
/** Max Budget */
max_budget?: number | null;
/** Soft Budget */
soft_budget?: number | null;
};
/** AccessGroupBudgetResponse */
AccessGroupBudgetResponse: {
/** Access Group */
access_group: string;
budget?: components["schemas"]["AccessGroupBudget"] | null;
/** Spend */
spend: number;
};
/** AccessGroupCreateRequest */
AccessGroupCreateRequest: {
/** Access Agent Ids */
access_agent_ids?: string[] | null;
/** Access Group Name */
access_group_name: string;
/** Access Mcp Server Ids */
access_mcp_server_ids?: string[] | null;
/** Access Model Names */
access_model_names?: string[] | null;
/** Access Skill Names */
access_skill_names?: string[] | null;
/** Assigned Key Ids */
assigned_key_ids?: string[] | null;
/** Assigned Team Ids */
assigned_team_ids?: string[] | null;
/** Description */
description?: string | null;
/** Mcp Tool Permissions */
mcp_tool_permissions?: {
[key: string]: string[];
} | null;
};
/** AccessGroupInfo */
AccessGroupInfo: {
/** Access Group */
access_group: string;
budget?: components["schemas"]["AccessGroupBudget"] | null;
/** Deployment Count */
deployment_count: number;
/** Model Names */
model_names: string[];
/** Spend */
spend?: number | null;
};
/**
* AccessGroupResource
* @description A resource referenced by an access group. `name` is null when the id no longer resolves or has no alias.
*/
AccessGroupResource: {
/** Id */
id: string;
/** Name */
name: string | null;
};
/** AccessGroupResponse */
AccessGroupResponse: {
/** Access Agent Ids */
access_agent_ids: string[];
/** Access Agents */
access_agents: components["schemas"]["AccessGroupResource"][];
/** Access Group Id */
access_group_id: string;
/** Access Group Name */
access_group_name: string;
/** Access Mcp Server Ids */
access_mcp_server_ids: string[];
/** Access Mcp Servers */
access_mcp_servers: components["schemas"]["AccessGroupResource"][];
/** Access Model Names */
access_model_names: string[];
/**
* Access Skill Names
* @default []
*/
access_skill_names: string[];
/** Assigned Key Ids */
assigned_key_ids: string[];
/** Assigned Keys */
assigned_keys: components["schemas"]["AccessGroupResource"][];
/** Assigned Team Ids */
assigned_team_ids: string[];
/** Assigned Teams */
assigned_teams: components["schemas"]["AccessGroupResource"][];
/**
* Created At
* Format: date-time
*/
created_at: string;
/** Created By */
created_by?: string | null;
/** Description */
description?: string | null;
/** Mcp Tool Permissions */
mcp_tool_permissions?: {
[key: string]: string[];
} | null;
/**
* Updated At
* Format: date-time
*/
updated_at: string;
/** Updated By */
updated_by?: string | null;
};
/** AccessGroupUpdateRequest */
AccessGroupUpdateRequest: {
/** Access Agent Ids */
access_agent_ids?: string[] | null;
/** Access Group Name */
access_group_name?: string | null;
/** Access Mcp Server Ids */
access_mcp_server_ids?: string[] | null;
/** Access Model Names */
access_model_names?: string[] | null;
/** Access Skill Names */
access_skill_names?: string[] | null;
/** Assigned Key Ids */
assigned_key_ids?: string[] | null;
/** Assigned Team Ids */
assigned_team_ids?: string[] | null;
/** Description */
description?: string | null;
/** Mcp Tool Permissions */
mcp_tool_permissions?: {
[key: string]: string[];
} | null;
};
/**
* AcknowledgedSafetyCheck
* @description A pending safety check for the computer call.
*/
AcknowledgedSafetyCheck: {
/** Code */
code?: string | null;
/** Id */
id: string;
/** Message */
message?: string | null;
} & {
[key: string]: unknown;
};
/**
* Action
* @description The shell commands and limits that describe how to run the tool call.
*/
Action: {
/** Commands */
commands: string[];
/** Max Output Length */
max_output_length?: number | null;
/** Timeout Ms */
timeout_ms?: number | null;
} & {
[key: string]: unknown;
};
/**
* ActionClick
* @description A click action.
*/
ActionClick: {
/**
* Button
* @enum {string}
*/
button: "left" | "right" | "wheel" | "back" | "forward";
/** Keys */
keys?: string[] | null;
/**
* Type
* @constant
*/
type: "click";
/** X */
x: number;
/** Y */
y: number;
} & {
[key: string]: unknown;
};
/**
* ActionDoubleClick
* @description A double click action.
*/
ActionDoubleClick: {
/** Keys */
keys?: string[] | null;
/**
* Type
* @constant
*/
type: "double_click";
/** X */
x: number;
/** Y */
y: number;
} & {
[key: string]: unknown;
};
/**
* ActionDrag
* @description A drag action.
*/
ActionDrag: {
/** Keys */
keys?: string[] | null;
/** Path */
path: components["schemas"]["ActionDragPath"][];
/**
* Type
* @constant
*/
type: "drag";
} & {
[key: string]: unknown;
};
/**
* ActionDragPath
* @description An x/y coordinate pair, e.g. `{ x: 100, y: 200 }`.
*/
ActionDragPath: {
/** X */
x: number;
/** Y */
y: number;
} & {
[key: string]: unknown;
};
/**
* ActionFind
* @description Action type "find_in_page": Searches for a pattern within a loaded page.
*/
ActionFind: {
/** Pattern */
pattern: string;
/**
* Type
* @constant
*/
type: "find_in_page";
/** Url */
url: string;
} & {
[key: string]: unknown;
};
/**
* ActionKeypress
* @description A collection of keypresses the model would like to perform.
*/
ActionKeypress: {
/** Keys */
keys: string[];
/**
* Type
* @constant
*/
type: "keypress";
} & {
[key: string]: unknown;
};
/**
* ActionMove
* @description A mouse move action.
*/
ActionMove: {
/** Keys */
keys?: string[] | null;
/**
* Type
* @constant
*/
type: "move";
/** X */
x: number;
/** Y */
y: number;
} & {
[key: string]: unknown;
};
/**
* ActionOpenPage
* @description Action type "open_page" - Opens a specific URL from search results.
*/
ActionOpenPage: {
/**
* Type
* @constant
*/
type: "open_page";
/** Url */
url?: string | null;
} & {
[key: string]: unknown;
};
/**
* ActionScreenshot
* @description A screenshot action.
*/
ActionScreenshot: {
/**
* Type
* @constant
*/
type: "screenshot";
} & {
[key: string]: unknown;
};
/**
* ActionScroll
* @description A scroll action.
*/
ActionScroll: {
/** Keys */
keys?: string[] | null;
/** Scroll X */
scroll_x: number;
/** Scroll Y */
scroll_y: number;
/**
* Type
* @constant
*/
type: "scroll";
/** X */
x: number;
/** Y */
y: number;
} & {
[key: string]: unknown;
};
/**
* ActionSearch
* @description Action type "search" - Performs a web search query.
*/
ActionSearch: {
/** Queries */
queries?: string[] | null;
/** Query */
query: string;
/** Sources */
sources?: components["schemas"]["ActionSearchSource"][] | null;
/**
* Type
* @constant
*/
type: "search";
} & {
[key: string]: unknown;
};
/**
* ActionSearchSource
* @description A source used in the search.
*/
ActionSearchSource: {
/**
* Type
* @constant
*/
type: "url";
/** Url */
url: string;
} & {
[key: string]: unknown;
};
/**
* ActionType
* @description An action to type in text.
*/
ActionType: {
/** Text */
text: string;
/**
* Type
* @constant
*/
type: "type";
} & {
[key: string]: unknown;
};
/**
* ActionWait
* @description A wait action.
*/
ActionWait: {
/**
* Type
* @constant
*/
type: "wait";
} & {
[key: string]: unknown;
};
/**
* ActiveUsersAnalyticsResponse
* @description Response for active users analytics
*/
ActiveUsersAnalyticsResponse: {
/** Results */
results: components["schemas"]["TagActiveUsersResponse"][];
};
/** AdaptiveRouterWeights */
AdaptiveRouterWeights: {
/**
* Cost
* @default 0.3
*/
cost: number;
/**
* Quality
* @default 0.7
*/
quality: number;
};
/** AddTeamCallback */
AddTeamCallback: {
/** Callback Name */
callback_name: string;
/**
* Callback Type
* @default success_and_failure
*/
callback_type: ("success" | "failure" | "success_and_failure") | null;
/** Callback Vars */
callback_vars: {
[key: string]: string;
};
};
/** AdminIgnore */
AdminIgnore: {
/**
* Ignored At
* Format: date-time
*/
ignored_at: string;
/** Ignored By */
ignored_by: string;
/**
* Ignored Until
* @description When the snooze ends. Null means the warning is ignored until an admin restores it.
*/
ignored_until?: string | null;
/**
* @description discriminator enum property added by openapi-typescript
* @enum {string}
*/
source: "admin";
};
/**
* AgentCapabilities
* @description Defines optional capabilities supported by an agent.
*/
AgentCapabilities: {
/** Extensions */
extensions?: components["schemas"]["AgentExtension"][] | null;
/** Pushnotifications */
pushNotifications?: boolean | null;
/** Statetransitionhistory */
stateTransitionHistory?: boolean | null;
/** Streaming */
streaming?: boolean | null;
};
/**
* AgentCard
* @description The AgentCard is a self-describing manifest for an agent.
* It provides essential metadata including the agent's identity, capabilities,
* skills, supported communication methods, and security requirements.
*/
AgentCard: {
/** Additionalinterfaces */
additionalInterfaces?: components["schemas"]["AgentInterface"][] | null;
capabilities?: components["schemas"]["AgentCapabilities"];
/** Defaultinputmodes */
defaultInputModes?: string[];
/** Defaultoutputmodes */
defaultOutputModes?: string[];
/** Description */
description?: string;
/** Documentationurl */
documentationUrl?: string | null;
/** Iconurl */
iconUrl?: string | null;
/** Name */
name?: string;
/** Preferredtransport */
preferredTransport?: string | null;
/** Protocolversion */
protocolVersion?: string;
provider?: components["schemas"]["AgentProvider"] | null;
/** Security */
security?: {
[key: string]: string[];
}[] | null;
/** Securityschemes */
securitySchemes?: {
[key: string]: components["schemas"]["APIKeySecurityScheme"] | components["schemas"]["HTTPAuthSecurityScheme"] | components["schemas"]["OAuth2SecurityScheme"] | components["schemas"]["OpenIdConnectSecurityScheme"] | components["schemas"]["MutualTLSSecurityScheme"];
} | null;
/** Signatures */
signatures?: components["schemas"]["AgentCardSignature"][] | null;
/** Skills */
skills?: components["schemas"]["AgentSkill"][];
/** Supportsauthenticatedextendedcard */
supportsAuthenticatedExtendedCard?: boolean | null;
/** Url */
url?: string;
/** Version */
version?: string;
};
/**
* AgentCardSignature
* @description Represents a JWS signature of an AgentCard.
*/
AgentCardSignature: {
/** Header */
header?: {
[key: string]: unknown;
} | null;
/** Protected */
protected?: string;
/** Signature */
signature?: string;
};
/** AgentConfig */
AgentConfig: {
/** Access Group Ids */
access_group_ids?: string[] | null;
agent_card_params?: components["schemas"]["AgentCard"];
/** Agent Name */
agent_name: string;
/** Enabled */
enabled?: boolean;
/**
* Execution Mode
* @enum {string}
*/
execution_mode?: "autonomous" | "delegated" | "both";
/** Extra Headers */
extra_headers?: string[] | null;
identity?: components["schemas"]["EntraIdentityConfig"] | null;
kill_switch?: components["schemas"]["AgentKillSwitchConfig"] | null;
/** Litellm Params */
litellm_params?: {
[key: string]: unknown;
};
object_permission?: components["schemas"]["AgentObjectPermission"];
/** Rpm Limit */
rpm_limit?: number | null;
/** Session Rpm Limit */
session_rpm_limit?: number | null;
/** Session Tpm Limit */
session_tpm_limit?: number | null;
/** Static Headers */
static_headers?: {
[key: string]: string;
} | null;
/** Tpm Limit */
tpm_limit?: number | null;
};
/** AgentCreateInfo */
AgentCreateInfo: {
/** Agent Type */
agent_type: string;
/** Agent Type Display Name */
agent_type_display_name: string;
/** Credential Fields */
credential_fields: components["schemas"]["AgentCredentialField"][];
/** Description */
description?: string | null;
/** Litellm Params Template */
litellm_params_template?: {
[key: string]: string;
} | null;
/** Logo Url */
logo_url?: string | null;
/** Model Template */
model_template?: string | null;
};
/** AgentCredentialField */
AgentCredentialField: {
/** Default Value */
default_value?: string | null;
/**
* Field Type
* @default text
* @enum {string}
*/
field_type: "text" | "password" | "select" | "upload" | "textarea";
/** Include In Litellm Params */
include_in_litellm_params?: boolean | null;
/** Key */
key: string;
/** Label */
label: string;
/** Options */
options?: string[] | null;
/** Placeholder */
placeholder?: string | null;
/**
* Required
* @default false
*/
required: boolean;
/** Tooltip */
tooltip?: string | null;
/** Validation Message */
validation_message?: string | null;
/** Validation Pattern */
validation_pattern?: string | null;
};
/**
* AgentExtension
* @description A declaration of a protocol extension supported by an Agent.
*/
AgentExtension: {
/** Description */
description?: string | null;
/** Params */
params?: {
[key: string]: unknown;
} | null;
/** Required */
required?: boolean | null;
/** Uri */
uri?: string;
};
/** AgentIdentityBinding */
AgentIdentityBinding: {
/**
* Active
* @default true
*/
active: boolean;
/** Agent Id */
agent_id: string;
/** Client Id */
client_id: string;
/** Issuer */
issuer: string;
/** Last Authenticated At */
last_authenticated_at?: string | null;
/**
* Provider
* @constant
*/
provider: "microsoft_entra";
/**
* Required Roles
* @default []
*/
required_roles: string[];
/**
* Required Scopes
* @default [
* "user_impersonation"
* ]
*/
required_scopes: string[];
/** Revision */
revision: string;
/** Service Principal Id */
service_principal_id?: string | null;
/** Tenant Id */
tenant_id: string;
};
/**
* AgentInterface
* @description Declares a combination of a target URL and a transport protocol.
*/
AgentInterface: {
/** Transport */
transport?: string;
/** Url */
url?: string;
};
/** AgentKeySummary */
AgentKeySummary: {
/** Key Alias */
key_alias?: string | null;
/** Key Name */
key_name?: string | null;
/** Token */
token: string;
};
/** AgentKillSwitchApiKeyAuth */
AgentKillSwitchApiKeyAuth: {
/** Api Key */
api_key: string;
/**
* Header Name
* @default x-api-key
*/
header_name: string;
/**
* @description discriminator enum property added by openapi-typescript
* @enum {string}
*/
type: "api_key";
};
/** AgentKillSwitchBasicAuth */
AgentKillSwitchBasicAuth: {
/** Password */
password: string;
/**
* @description discriminator enum property added by openapi-typescript
* @enum {string}
*/
type: "basic";
/** Username */
username: string;
};
/** AgentKillSwitchBearerAuth */
AgentKillSwitchBearerAuth: {
/** Token */
token: string;
/**
* @description discriminator enum property added by openapi-typescript
* @enum {string}
*/
type: "bearer";
};
/**
* AgentKillSwitchConfig
* @description Webhook an admin fires to shut an agent down out of band. LiteLLM only
* makes the call; whatever the endpoint does with it is the agent's business.
*/
AgentKillSwitchConfig: {
/** Auth */
auth?: (components["schemas"]["AgentKillSwitchBearerAuth"] | components["schemas"]["AgentKillSwitchApiKeyAuth"] | components["schemas"]["AgentKillSwitchBasicAuth"]) | null;
/** Body */
body?: {
[key: string]: unknown;
} | null;
/** Headers */
headers?: {
[key: string]: string;
};
/**
* Method
* @default POST
* @enum {string}
*/
method: "POST" | "PUT" | "PATCH" | "DELETE" | "GET";
/** Query Params */
query_params?: {
[key: string]: string;
};
/** Url */
url: string;
};
/** AgentKillSwitchResult */
AgentKillSwitchResult: {
/** Agent Id */
agent_id: string;
/** Error */
error?: string | null;
/**
* Method
* @enum {string}
*/
method: "POST" | "PUT" | "PATCH" | "DELETE" | "GET";
/** Response Body */
response_body?: string | null;
/** Status Code */
status_code?: number | null;
/** Url */
url: string;
};
/** AgentMakePublicResponse */
AgentMakePublicResponse: {
/** Message */
message: string;
/** Public Agent Groups */
public_agent_groups: string[];
/** Updated By */
updated_by: string;
};
/** AgentNode */
AgentNode: {
/** Duration Ms */
duration_ms: number;
/** Invocations */
invocations: number;
/** Llm Calls */
llm_calls: number;
/** Name */
name: string;
/** Parent Agent */
parent_agent: string | null;
/** Priced Calls */
priced_calls: number;
/** Spend */
spend: number | null;
/** Tool Calls */
tool_calls: number;
};
/** AgentObjectPermission */
AgentObjectPermission: {
/** Agents */
agents?: string[] | null;
/** Mcp Access Groups */
mcp_access_groups?: string[] | null;
/** Mcp Servers */
mcp_servers?: string[] | null;
/** Mcp Tool Permissions */
mcp_tool_permissions?: {
[key: string]: string[];
} | null;
/** Mcp Toolsets */
mcp_toolsets?: string[] | null;
/** Models */
models?: string[] | null;
};
/**
* AgentProvider
* @description Represents the service provider of an agent.
*/
AgentProvider: {
/** Organization */
organization?: string;
/** Url */
url?: string;
};
/** AgentResponse */
AgentResponse: {
/** Access Group Ids */
access_group_ids?: string[] | null;
/** Agent Card Params */
agent_card_params: {
[key: string]: unknown;
};
/** Agent Id */
agent_id: string;
/** Agent Name */
agent_name: string;
/** Created At */
created_at?: string | null;
/** Created By */
created_by?: string | null;
/**
* Enabled
* @default true
*/
enabled: boolean;
/**
* Execution Mode
* @default autonomous
* @enum {string}
*/
execution_mode: "autonomous" | "delegated" | "both";
/** Extra Headers */
extra_headers?: string[] | null;
identity?: components["schemas"]["AgentIdentityBinding"] | null;
/**
* Identity Managed
* @default false
*/
identity_managed: boolean;
/**
* Jwt Auth Configured
* @default false
*/
jwt_auth_configured: boolean;
/** Keys */
keys?: components["schemas"]["AgentKeySummary"][] | null;
kill_switch?: components["schemas"]["AgentKillSwitchConfig"] | null;
/** Litellm Params */
litellm_params?: {
[key: string]: unknown;
} | null;
/** Object Permission */
object_permission?: {
[key: string]: unknown;
} | null;
/** Rpm Limit */
rpm_limit?: number | null;
/** Search Score */
search_score?: number | null;
/** Session Rpm Limit */
session_rpm_limit?: number | null;
/** Session Tpm Limit */
session_tpm_limit?: number | null;
/** Spend */
spend?: number | null;
/** Static Headers */
static_headers?: {
[key: string]: string;
} | null;
/** Tpm Limit */
tpm_limit?: number | null;
/** Updated At */
updated_at?: string | null;
/** Updated By */
updated_by?: string | null;
};
/**
* AgentSkill
* @description Represents a distinct capability or function that an agent can perform.
*/
AgentSkill: {
/** Description */
description?: string;
/** Examples */
examples?: string[] | null;
/** Id */
id?: string;
/** Inputmodes */
inputModes?: string[] | null;
/** Name */
name?: string;
/** Outputmodes */
outputModes?: string[] | null;
/** Security */
security?: {
[key: string]: string[];
}[] | null;
/** Tags */
tags?: string[];
};
/** AgentSkillsIndex */
AgentSkillsIndex: {
/**
* $Schema
* @default https://schemas.agentskills.io/discovery/0.2.0/schema.json
*/
$schema: string;
/** Skills */
skills: components["schemas"]["AgentSkillsIndexEntry"][];
};
/** AgentSkillsIndexEntry */
AgentSkillsIndexEntry: {
/** Description */
description: string;
/** Digest */
digest: string;
/** Name */
name: string;
/**
* Type
* @constant
*/
type: "archive";
/** Url */
url: string;
};
/**
* AlertType
* @description Enum for alert types and management event types
* @enum {string}
*/
AlertType: "llm_exceptions" | "llm_too_slow" | "llm_requests_hanging" | "budget_alerts" | "spend_reports" | "failed_tracking_spend" | "user_spend_thresholds" | "user_spend_anomalies" | "db_exceptions" | "daily_reports" | "cooldown_deployment" | "new_model_added" | "model_deprecation_warnings" | "outage_alerts" | "region_outage_alerts" | "fallback_reports" | "new_virtual_key_created" | "virtual_key_updated" | "virtual_key_deleted" | "new_team_created" | "team_updated" | "team_deleted" | "new_internal_user_created" | "internal_user_updated" | "internal_user_deleted" | "mcp_tool_description_blocked" | "mcp_pinned_tools_changed";
/** AllowedVectorStoreIndexItem */
AllowedVectorStoreIndexItem: {
/** Index Name */
index_name: string;
/** Index Permissions */
index_permissions: ("read" | "write")[];
};
/**
* AnnotationContainerFileCitation
* @description A citation for a container file used to generate a model response.
*/
AnnotationContainerFileCitation: {
/** Container Id */
container_id: string;
/** End Index */
end_index: number;
/** File Id */
file_id: string;
/** Filename */
filename: string;
/** Start Index */
start_index: number;
/**
* Type
* @constant
*/
type: "container_file_citation";
} & {
[key: string]: unknown;
};
/**
* AnnotationFileCitation
* @description A citation to a file.
*/
AnnotationFileCitation: {
/** File Id */
file_id: string;
/** Filename */
filename: string;
/** Index */
index: number;
/**
* Type
* @constant
*/
type: "file_citation";
} & {
[key: string]: unknown;
};
/**
* AnnotationFilePath
* @description A path to a file.
*/
AnnotationFilePath: {
/** File Id */
file_id: string;
/** Index */
index: number;
/**
* Type
* @constant
*/
type: "file_path";
} & {
[key: string]: unknown;
};
/**
* AnnotationURLCitation
* @description A citation for a web resource used to generate a model response.
*/
AnnotationURLCitation: {
/** End Index */
end_index: number;
/** Start Index */
start_index: number;
/** Title */
title: string;
/**
* Type
* @constant
*/
type: "url_citation";
/** Url */
url: string;
} & {
[key: string]: unknown;
};
/** ApplyGuardrailRequest */
ApplyGuardrailRequest: {
/** Entities */
entities?: components["schemas"]["PiiEntityType"][] | null;
/** Guardrail Name */
guardrail_name: string;
/**
* Input Type
* @default request
*/
input_type: string;
/** Language */
language?: string | null;
/** Messages */
messages?: {
[key: string]: unknown;
}[] | null;
/** Metadata */
metadata?: {
[key: string]: unknown;
} | null;
/** Text */
text: string;
};
/** ApplyGuardrailResponse */
ApplyGuardrailResponse: {
/** Response Text */
response_text: string;
};
/**
* ApplyPatchCall
* @description A tool call representing a request to create, delete, or update files using diff patches.
*/
ApplyPatchCall: {
/** Call Id */
call_id: string;
/** Id */
id?: string | null;
/** Operation */
operation: components["schemas"]["ApplyPatchCallOperationCreateFile"] | components["schemas"]["ApplyPatchCallOperationDeleteFile"] | components["schemas"]["ApplyPatchCallOperationUpdateFile"];
/**
* Status
* @enum {string}
*/
status: "in_progress" | "completed";
/**
* Type
* @constant
*/
type: "apply_patch_call";
};
/**
* ApplyPatchCallOperationCreateFile
* @description Instruction for creating a new file via the apply_patch tool.
*/
ApplyPatchCallOperationCreateFile: {
/** Diff */
diff: string;
/** Path */
path: string;
/**
* Type
* @constant
*/
type: "create_file";
};
/**
* ApplyPatchCallOperationDeleteFile
* @description Instruction for deleting an existing file via the apply_patch tool.
*/
ApplyPatchCallOperationDeleteFile: {
/** Path */
path: string;
/**
* Type
* @constant
*/
type: "delete_file";
};
/**
* ApplyPatchCallOperationUpdateFile
* @description Instruction for updating an existing file via the apply_patch tool.
*/
ApplyPatchCallOperationUpdateFile: {
/** Diff */
diff: string;
/** Path */
path: string;
/**
* Type
* @constant
*/
type: "update_file";
};
/**
* ApplyPatchCallOutput
* @description The streamed output emitted by an apply patch tool call.
*/
ApplyPatchCallOutput: {
/** Call Id */
call_id: string;
/** Id */
id?: string | null;
/** Output */
output?: string | null;
/**
* Status
* @enum {string}
*/
status: "completed" | "failed";
/**
* Type
* @constant
*/
type: "apply_patch_call_output";
};
/**
* ApplyPatchTool
* @description Allows the assistant to create, delete, or update files using unified diffs.
*/
ApplyPatchTool: {
/**
* Type
* @constant
*/
type: "apply_patch";
} & {
[key: string]: unknown;
};
/**
* ApplyPatchToolParam
* @description Allows the assistant to create, delete, or update files using unified diffs.
*/
ApplyPatchToolParam: {
/**
* Type
* @constant
*/
type: "apply_patch";
};
/**
* AttachmentImpactResponse
* @description Response for estimating the impact of a policy attachment.
*/
AttachmentImpactResponse: {
/**
* Affected Keys Count
* @description Number of keys that would be affected (named + unnamed).
* @default 0
*/
affected_keys_count: number;
/**
* Affected Teams Count
* @description Number of teams that would be affected (named + unnamed).
* @default 0
*/
affected_teams_count: number;
/**
* Sample Keys
* @description Sample of affected key aliases (up to 10).
*/
sample_keys?: string[];
/**
* Sample Teams
* @description Sample of affected team aliases (up to 10).
*/
sample_teams?: string[];
/**
* Unnamed Keys Count
* @description Number of affected keys without an alias.
* @default 0
*/
unnamed_keys_count: number;
/**
* Unnamed Teams Count
* @description Number of affected teams without an alias.
* @default 0
*/
unnamed_teams_count: number;
};
/**
* AuditLogResponse
* @description Response model for a single audit log entry
*/
AuditLogResponse: {
/** Action */
action: string;
/** Before Value */
before_value?: {
[key: string]: unknown;
} | null;
/** Changed By */
changed_by: string;
/** Changed By Api Key */
changed_by_api_key: string;
/** Id */
id: string;
/** Object Id */
object_id: string;
/** Table Name */
table_name: string;
/**
* Updated At
* Format: date-time
*/
updated_at: string;
/** Updated Values */
updated_values?: {
[key: string]: unknown;
} | null;
};
/** AutoRouterAllowance */
AutoRouterAllowance: {
/**
* Available
* @default true
*/
available: boolean;
/** Key */
key: string;
/** Limit */
limit: number | null;
/** Remaining */
remaining: number | null;
/**
* Used By This Router
* @default false
*/
used_by_this_router: boolean;
};
/** AutoRouterAvailabilityRequest */
AutoRouterAvailabilityRequest: {
/** Complexity Router Config */
complexity_router_config?: {
[key: string]: unknown;
} | null;
/** Saved Model Id */
saved_model_id?: string | null;
/** Team Id */
team_id?: string | null;
};
/** AutoRouterAvailabilityResponse */
AutoRouterAvailabilityResponse: {
/** Allowances */
allowances: components["schemas"]["AutoRouterAllowance"][];
/** Error */
error?: string | null;
};
/**
* AutoRouterBenchmarkGroup
* @description One auto-router's slice of the benchmarks.
*/
AutoRouterBenchmarkGroup: {
/**
* Avg Session Seconds
* @description Lifetime seconds per overlapping session; null as above
*/
avg_session_seconds: number | null;
/**
* Avg Tokens Per Session
* @description Lifetime tokens per overlapping session; null as above
*/
avg_tokens_per_session: number | null;
/**
* Avg Turns Per Session
* @description Lifetime turns per overlapping session; null when the window has routed requests but no session rows for this router type, such as an alias whose router type changed mid-session
*/
avg_turns_per_session: number | null;
/**
* Baseline Spend
* @description Estimated single-model cost: compared actual spend plus recorded savings; null when traffic has no recorded savings
*/
baseline_spend: number | null;
cache: components["schemas"]["AutoRouterCacheStats"];
/**
* Classifier Cost
* @description Recorded LLM classifier cost already included in spend; null when any session turns predate subtotal recording, and zero for an empty window
*/
classifier_cost: number | null;
/**
* Router Name
* @description The auto-router alias requests were sent to
*/
router_name: string;
/**
* Router Type
* @description complexity, adaptive or quality
*/
router_type: string;
/**
* Saved Pct
* @description Recorded savings over baseline_spend, as a percentage
*/
saved_pct: number | null;
/**
* Saved Spend
* @description Recorded savings on the selected UTC days; null when traffic has no recorded savings estimates. On totals this is the same daily figure the Overall savings view reports
*/
saved_spend: number | null;
/**
* Savings Estimated Actual Spend
* @description Actual spend, including classifier cost, for the compared requests
*/
savings_estimated_actual_spend: number;
/**
* Savings Estimated Classifier Cost
* @description Classifier cost included in the compared actual spend; null when classification costs for those requests are unavailable
*/
savings_estimated_classifier_cost?: number | null;
/**
* Savings Estimated Turns
* @description Requests compared against the baseline: every request on complexity routers that recorded savings
*/
savings_estimated_turns: number;
/**
* Sessions
* @description Sessions overlapping the window, counted whole
*/
sessions: number;
/**
* Spend
* @description What the selected days' routed traffic actually cost
*/
spend: number;
/**
* Tier Turns
* @description Turns per tier, keyed by the tier name the routing decision recorded at request time (never re-derived at read time, since the tier-to-model mapping is mutable config). Tier names are scoped to this group's router_type and are not comparable across types: a complexity router reports 'SIMPLE'/'MEDIUM'/'COMPLEX'/'REASONING', a quality router reports its numeric quality tier, and an adaptive router records no tier at all. Turns no tier served (the classifier fell back to default_model) are absent rather than pooled under a sentinel key, so the values may sum to less than turns
*/
tier_turns?: {
[key: string]: number;
};
/**
* Total Tokens
* @description Input and output tokens of routed generation requests on the selected UTC days, excluding classifier tokens; null when any selected requests predate daily token recording
*/
total_tokens?: number | null;
/**
* Turns
* @description Auto-routed requests on the selected UTC days
*/
turns: number;
/**
* Unattributed Saved Spend
* @description Part of saved_spend no router's daily rows account for, such as history recorded before per-router daily tracking; when set, baseline_spend and saved_pct are null
*/
unattributed_saved_spend?: number | null;
};
/**
* AutoRouterBenchmarkTotals
* @description Auto-routed traffic in the window. Turns, spend and savings count requests on the selected UTC days;
* the session averages and cache stats describe every session overlapping the window, whole.
*/
AutoRouterBenchmarkTotals: {
/**
* Avg Session Seconds
* @description Lifetime seconds per overlapping session; null as above
*/
avg_session_seconds: number | null;
/**
* Avg Tokens Per Session
* @description Lifetime tokens per overlapping session; null as above
*/
avg_tokens_per_session: number | null;
/**
* Avg Turns Per Session
* @description Lifetime turns per overlapping session; null when the window has routed requests but no session rows for this router type, such as an alias whose router type changed mid-session
*/
avg_turns_per_session: number | null;
/**
* Baseline Spend
* @description Estimated single-model cost: compared actual spend plus recorded savings; null when traffic has no recorded savings
*/
baseline_spend: number | null;
cache: components["schemas"]["AutoRouterCacheStats"];
/**
* Classifier Cost
* @description Recorded LLM classifier cost already included in spend; null when any session turns predate subtotal recording, and zero for an empty window
*/
classifier_cost: number | null;
/**
* Saved Pct
* @description Recorded savings over baseline_spend, as a percentage
*/
saved_pct: number | null;
/**
* Saved Spend
* @description Recorded savings on the selected UTC days; null when traffic has no recorded savings estimates. On totals this is the same daily figure the Overall savings view reports
*/
saved_spend: number | null;
/**
* Savings Estimated Actual Spend
* @description Actual spend, including classifier cost, for the compared requests
*/
savings_estimated_actual_spend: number;
/**
* Savings Estimated Classifier Cost
* @description Classifier cost included in the compared actual spend; null when classification costs for those requests are unavailable
*/
savings_estimated_classifier_cost?: number | null;
/**
* Savings Estimated Turns
* @description Requests compared against the baseline: every request on complexity routers that recorded savings
*/
savings_estimated_turns: number;
/**
* Sessions
* @description Sessions overlapping the window, counted whole
*/
sessions: number;
/**
* Spend
* @description What the selected days' routed traffic actually cost
*/
spend: number;
/**
* Total Tokens
* @description Input and output tokens of routed generation requests on the selected UTC days, excluding classifier tokens; null when any selected requests predate daily token recording
*/
total_tokens?: number | null;
/**
* Turns
* @description Auto-routed requests on the selected UTC days
*/
turns: number;
/**
* Unattributed Saved Spend
* @description Part of saved_spend no router's daily rows account for, such as history recorded before per-router daily tracking; when set, baseline_spend and saved_pct are null
*/
unattributed_saved_spend?: number | null;
};
/**
* AutoRouterBenchmarksResponse
* @description Benchmarks for the auto-router dashboard, aggregated from the per-session and per-day rollups.
*/
AutoRouterBenchmarksResponse: {
/**
* End Date
* @description Window end day, YYYY-MM-DD UTC, inclusive
*/
end_date: string;
/**
* Groups
* @description One entry per auto-router, listed from the model registry rather than from the rollup, so a router appears as soon as it is configured and reads zero until it serves traffic. Semantic auto-routers are absent: they record no routing decision, so no session can ever be attributed to them
*/
groups: components["schemas"]["AutoRouterBenchmarkGroup"][];
/**
* Routers In Scope
* @description How many groups this response carries. Every auto-router configured on the proxy counts, whether or not it served anything in the window. To count only the routers that did serve traffic, filter `groups` to the entries whose `sessions` is above zero
*/
routers_in_scope: number;
/**
* Start Date
* @description Window start day, YYYY-MM-DD UTC, inclusive
*/
start_date: string;
totals: components["schemas"]["AutoRouterBenchmarkTotals"];
};
/**
* AutoRouterCacheBucket
* @description One prompt-caching bucket of turns, with how often those turns hit the cache.
*/
AutoRouterCacheBucket: {
/**
* Hit Rate Pct
* @description hits over this bucket's turns, as a percentage
*/
hit_rate_pct: number;
/**
* Hits
* @description Turns in this bucket whose response reported cache-read tokens
*/
hits: number;
/**
* Turns
* @description Turns classified into this bucket
*/
turns: number;
};
/**
* AutoRouterCacheStats
* @description Prompt-caching behaviour of auto-routed turns, bucketed by what the router did.
*
* Every in-order turn falls in exactly one bucket: the session stayed on the same model,
* visited a model for the first time (cold by design), or returned to a model it had
* already used. Out-of-order turns (cross-pod flush races) are counted but not bucketed.
*/
AutoRouterCacheStats: {
/**
* Coverage Pct
* @description Share of turns that carried cache telemetry
*/
coverage_pct: number;
first_visit: components["schemas"]["AutoRouterCacheBucket"];
/**
* Hit Rate Pct
* @description All cache hits over telemetry-bearing turns
*/
hit_rate_pct: number;
/**
* Return Misses Expired
* @description Return-to-tier misses where the model's recorded cache TTL had lapsed
*/
return_misses_expired: number;
/**
* Return Misses Unknown
* @description Return-to-tier misses with no recorded TTL to attribute against
*/
return_misses_unknown: number;
/**
* Return Misses Within Ttl
* @description Return-to-tier misses inside the recorded TTL: the prefix changed or the provider evicted the entry early; billing telemetry cannot distinguish the two
*/
return_misses_within_ttl: number;
return_to_tier: components["schemas"]["AutoRouterCacheBucket"];
same_model: components["schemas"]["AutoRouterCacheBucket"];
/**
* Ttl 1H Turns
* @description Turns whose cache write used the one-hour TTL
*/
ttl_1h_turns: number;
/**
* Ttl 5M Turns
* @description Turns whose cache write used the five-minute TTL
*/
ttl_5m_turns: number;
/**
* Unordered Turns
* @description Turns that arrived out of order and were not bucketed
*/
unordered_turns: number;
};
/**
* AutoRouterClassifierDefaultPromptResponse
* @description The built-in system prompt an auto-router's LLM classifier uses when none is configured.
*
* Served so the dashboard's prompt editor prefills the rubric the proxy actually sends, rather than
* a copy in the frontend that drifts the moment the rubric is edited.
*/
AutoRouterClassifierDefaultPromptResponse: {
/** System Prompt */
system_prompt: string;
};
/**
* AutoRouterClassifierPromptPreviewRequest
* @description A POST rather than query params: the classification sections are the operator's own text,
* which must not reach access logs through a URL.
*/
AutoRouterClassifierPromptPreviewRequest: {
/** Classification Examples */
classification_examples?: string | null;
/** Classification Prompt */
classification_prompt?: string | null;
classification_rubric?: components["schemas"]["ClassificationRubric"] | null;
/**
* Context Window Size
* @default 3
*/
context_window_size: number;
/** Tier Definitions */
tier_definitions?: components["schemas"]["TierDefinition"][] | null;
/** Tier Labels */
tier_labels?: {
[key: string]: string;
} | null;
};
/**
* AutoRouterPresetConfig
* @description The complexity_router_config a preset prefills.
*
* Only tiers is validated, because every dashboard consumer dereferences it; everything else
* passes through verbatim with unknown fields kept (extra="allow"), so a catalog published after
* this proxy shipped still serves its new fields intact.
*/
AutoRouterPresetConfig: {
tiers: components["schemas"]["AutoRouterPresetTiers"];
} & {
[key: string]: unknown;
};
/**
* AutoRouterPresetRecord
* @description One auto-router preset as served to the dashboard's template picker.
*/
AutoRouterPresetRecord: {
complexity_router_config: components["schemas"]["AutoRouterPresetConfig"];
/** Description */
description: string;
/** Label */
label: string;
} & {
[key: string]: unknown;
};
/**
* AutoRouterPresetTiers
* @description Exactly the four built-in tiers the dashboard's preset prefill can apply.
*
* extra="forbid" on purpose: a tier name this dashboard cannot apply would grey out or crash the
* picker, so such a catalog is rejected wholesale and the bundled one serves instead.
*/
AutoRouterPresetTiers: {
/** Complex */
COMPLEX: string[];
/** Medium */
MEDIUM: string[];
/** Reasoning */
REASONING: string[];
/** Simple */
SIMPLE: string[];
};
/**
* AutoRouterRoutingTestRequest
* @description A single request to classify against a complexity-router config that need not be saved yet.
*
* Carries the same fields the serving path carries, so a dry run classifies what a real turn
* would classify. `messages`, `system` and `tools` are forwarded to the routing hook untranslated,
* which is why they are typed loosely: the hook reads whatever dialect the surface produced, and
* validating them against one surface's schema would reject the others.
*/
AutoRouterRoutingTestRequest: {
/** @description The complexity router config to route against, in the shape /model/new accepts */
complexity_router_config: components["schemas"]["RequestComplexityRouterConfig"];
/**
* Default Model
* @description Model to route to when no tier resolves, i.e. complexity_router_default_model
*/
default_model?: string | null;
/**
* Messages
* @description The full message list to route, exactly as the serving path would receive it. Mutually exclusive with prompt
*/
messages?: {
[key: string]: unknown;
}[] | null;
/**
* Prompt
* @description A single ask to route, as an end user would send it. Mutually exclusive with messages
*/
prompt?: string | null;
/**
* Router Name
* @description Name reported as the router in the routing decision. Display only
* @default auto_router_routing_test
*/
router_name: string;
/**
* Saved Model Id
* @description Test this saved deployment's server-side configuration instead of the supplied config and default model
*/
saved_model_id?: string | null;
/**
* System
* @description The top-level system prompt an Anthropic /v1/messages body carries beside its messages
*/
system?: string | {
[key: string]: unknown;
}[] | null;
/**
* Team Id
* @description Team the router is being created for. Required for a team admin, who may only test their own team's routers
*/
team_id?: string | null;
/**
* Tools
* @description The tool definitions the request advertises, which decide whether the plan-mode floor applies
*/
tools?: {
[key: string]: unknown;
}[] | null;
};
/**
* AutoRouterRoutingTestResponse
* @description Where one prompt would have been routed, and why.
*/
AutoRouterRoutingTestResponse: {
/**
* Routed Model
* @description The model group the router picked
*/
routed_model: string;
/**
* Routed Model Configured
* @description Whether routed_model is a model group available to the caller, scoped to team_id when given. Never confirms models the caller could not use
*/
routed_model_configured: boolean;
/** @description The decision record this request would have written to its log row */
routing_decision: components["schemas"]["StandardLoggingRoutingDecision"];
};
/**
* AutoRouterSessionResponse
* @description One auto-routed session as its own key sees it: what the last turn ran on, and what the session cost
* against the router's savings baseline (the priciest model in its hardest tier).
*/
AutoRouterSessionResponse: {
/**
* Baseline Model
* @description The savings baseline recorded by most session turns, including historical turns, recorded turn by turn, so it still names the counterfactual after the router is reconfigured or removed. None when no turn recorded one: rows from before the baseline was recorded, and adaptive and quality routers, which derive no baseline and so report no savings
*/
baseline_model: string | null;
/**
* Baseline Models
* @description Session turns recording each baseline model; more than one entry means the router's baseline changed mid-session; these counts do not imply savings coverage
*/
baseline_models: {
[key: string]: number;
};
/**
* Baseline Spend
* @description Estimated single-model cost: spend plus recorded savings
*/
baseline_spend: number | null;
/**
* Last Model
* @description The deployment model the most recent turn was routed to
*/
last_model: string;
/**
* Router Name
* @description The auto-router alias the session's requests were sent to
*/
router_name: string;
/**
* Router Type
* @description complexity, adaptive or quality
*/
router_type: string;
/**
* Saved Spend
* @description Recorded historical savings plus newer estimates, net of classifier cost
*/
saved_spend: number | null;
/**
* Savings Estimated Actual Spend
* @description Actual spend, including classifier cost, for requests whose estimate recorded its baseline cost
*/
savings_estimated_actual_spend: number;
/**
* Savings Estimated Baseline Spend
* @description Estimated single-model cost for requests whose estimate recorded its baseline cost
*/
savings_estimated_baseline_spend: number | null;
/**
* Savings Estimated Turns
* @description Requests whose savings estimate recorded its baseline cost
*/
savings_estimated_turns: number;
/** Session Id */
session_id: string;
/**
* Spend
* @description What the session's routed traffic actually cost, classifier calls included
*/
spend: number;
/**
* Turns
* @description Auto-routed turns the rollup has recorded for this session so far
*/
turns: number;
};
/** AwsSessionTag */
AwsSessionTag: {
/** Key */
Key: string;
/** Value */
Value: string;
} & {
[key: string]: unknown;
};
/** BaseLitellmParams */
BaseLitellmParams: {
/**
* Additional Provider Specific Params
* @description Additional provider-specific parameters for generic guardrail APIs
*/
additional_provider_specific_params?: {
[key: string]: unknown;
} | null;
/**
* Api Base
* @description Base URL for the guardrail service API
*/
api_base?: string | null;
/**
* Api Endpoint
* @description Optional custom API endpoint for Model Armor
*/
api_endpoint?: string | null;
/**
* Api Key
* @description API key for the guardrail service
*/
api_key?: string | null;
/**
* Blocked Words
* @description List of blocked words with individual actions
*/
blocked_words?: components["schemas"]["BlockedWord"][] | null;
/**
* Blocked Words File
* @description Path to YAML file containing blocked_words list
*/
blocked_words_file?: string | null;
/**
* Categories
* @description List of prebuilt categories to enable (harmful_*, bias_*)
*/
categories?: components["schemas"]["ContentFilterCategoryConfig"][] | null;
/** @description Threshold configuration for Lakera guardrail categories */
category_thresholds?: components["schemas"]["LakeraCategoryThresholds"] | null;
/**
* Credentials
* @description Path to Google Cloud credentials JSON file or JSON string
*/
credentials?: string | null;
/**
* Custom Code
* @description Python-like code containing the apply_guardrail function for custom guardrail logic
*/
custom_code?: string | null;
/**
* Default On
* @description Whether the guardrail is enabled by default
*/
default_on?: boolean | null;
/**
* Detect Secrets Config
* @description Configuration for detect-secrets guardrail
*/
detect_secrets_config?: {
[key: string]: unknown;
} | null;
/**
* End Session After N Fails
* @description For /v1/realtime sessions: automatically close the session after this many guardrail violations.
*/
end_session_after_n_fails?: number | null;
/**
* Experimental Use Latest Role Message Only
* @description When True, guardrails only receive the latest message for the relevant role (e.g., newest user input pre-call, newest assistant output post-call)
* @default false
*/
experimental_use_latest_role_message_only: boolean | null;
/**
* Extra Headers
* @description Header names to forward from the client request to the guardrail (e.g. x-request-id). Only these headers' values are sent; others may be omitted or sent as [present]. Used by generic_guardrail_api (similar to MCP extra_headers).
*/
extra_headers?: string[] | null;
/**
* Fail On Error
* @description Whether to fail the request if the guardrail encounters an error. Implemented by guardrail='model_armor', 'generic_guardrail_api' and 'crowdstrike_aidr'. True (default) raises the error. False logs a critical error and lets the request proceed, so only a valid guardrail response can block or modify it.
* @default true
*/
fail_on_error: boolean | null;
/**
* Forward Messages As Sent
* @description Only honored by guardrail='generic_guardrail_api'. When True, structured_messages reaches the guardrail exactly as the client sent it, so rows can carry provider-native content blocks and keys the OpenAI chat types do not declare. Default False sends OpenAI-format chat rows with undeclared keys removed.
* @default false
*/
forward_messages_as_sent: boolean | null;
/**
* Guard Name
* @description Name of the guardrail in guardrails.ai
*/
guard_name?: string | null;
/**
* Inspect Embeddings
* @description When True, the Aim and Cato Networks guardrails send /embeddings `input` to the vendor as user messages. Off by default because embedding input is documents being indexed, not a conversation.
*/
inspect_embeddings?: boolean | null;
/**
* Keyword Redaction Tag
* @description Tag to use for keyword redaction
*/
keyword_redaction_tag?: string | null;
/**
* Location
* @description Google Cloud location/region (e.g., us-central1)
*/
location?: string | null;
/**
* Logging Only Continue On Input Failure
* @description when True, a flagged or raising logging_only request scan is logged and the response is still scanned, so both verdicts land. Only applies to mode logging_only and is ignored when logging_only_scope is 'input' or 'output'.
*/
logging_only_continue_on_input_failure?: boolean | null;
/**
* Logging Only Scope
* @description which direction a logging_only scan observes: 'input' (request) or 'output' (response); unset scans both directions. Only applies to mode logging_only; pre_call/post_call on the same guardrail keep blocking.
*/
logging_only_scope?: ("input" | "output") | null;
/**
* Mask Request Content
* @description Will mask request content if guardrail makes any changes
*/
mask_request_content?: boolean | null;
/**
* Mask Response Content
* @description Will mask response content if guardrail makes any changes
*/
mask_response_content?: boolean | null;
/**
* Max Concurrent Decision Calls
* @description Maximum decisions calls in flight at once on the guardrail in each proxy worker. Implemented by guardrail='decision_model'.
*/
max_concurrent_decision_calls?: number | null;
/**
* Max Input Chars
* @description Character budget for each text sent to the guardrail's model. Implemented by guardrail='decision_model'.
*/
max_input_chars?: number | null;
/**
* Model
* @description Optional field if guardrail requires a 'model' parameter
*/
model?: string | null;
/**
* Num Retries
* @description How many more times to send a guardrail check whose connection timed out waiting for the answer or was dropped after the request went out. Each retry opens a fresh connection and gets the full timeout. Connect failures, HTTP error statuses and BLOCKED verdicts are not retried. Unset or 0 sends each check once, and the most allowed is 5. Implemented by guardrail='generic_guardrail_api', other guardrails ignore it.
*/
num_retries?: number | null;
/**
* On Sensitive Data
* @description Action to take when sensitive data is detected. 'block' raises an exception (default behavior). 'route' reroutes the request to the model specified in sensitive_data_route_to_model.
*/
on_sensitive_data?: ("block" | "route") | null;
/**
* On Violation
* @description For /v1/realtime sessions: 'warn' speaks the violation message and continues; 'end_session' speaks the message and closes the connection. For guardrail='mcp_security': 'block' rejects the request; 'alert' only logs a warning.
*/
on_violation?: ("warn" | "end_session" | "block" | "alert") | null;
/**
* Only Scan New Messages
* @description When True, the guardrail only scans messages that have not already been scanned earlier in the same session (identified by litellm_session_id / session_id). Message content is hashed per session and cached; only the diff (new or edited messages) is sent to the guardrail provider on follow-up calls. Falls back to a full scan when the request has no session id or the cache is unavailable. Intended for blocking/detection guardrails; not applied when mask_request_content is set.
* @default false
*/
only_scan_new_messages: boolean | null;
/**
* Pangea Input Recipe
* @description Recipe for input (LLM request)
*/
pangea_input_recipe?: string | null;
/**
* Pangea Output Recipe
* @description Recipe for output (LLM response)
*/
pangea_output_recipe?: string | null;
/**
* Pattern Redaction Format
* @description Format string for pattern redaction (use {pattern_name} placeholder)
*/
pattern_redaction_format?: string | null;
/**
* Patterns
* @description List of patterns (prebuilt or custom regex) to detect
*/
patterns?: components["schemas"]["ContentFilterPattern"][] | null;
/**
* Realtime Violation Message
* @description The message the bot speaks aloud when a /v1/realtime guardrail fires. Falls back to violation_message_template if not set.
*/
realtime_violation_message?: string | null;
/**
* Run In Parallel
* @description When True, this pre_call or post_call guardrail runs concurrently with other opted-in guardrails of the same hook, after the sequential guardrails have run. Use only for block-only guardrails that inspect and reject; do not enable it for guardrails that modify the request or response (e.g. PII masking or sensitive-data routing), since parallel runs share one snapshot and their mutations would race.
*/
run_in_parallel?: boolean | null;
/**
* Sanitize Error Detail
* @description For guardrail='model_armor': omit the raw Model Armor response from caller-facing errors and logs by default. Set False to restore verbose output.
* @default true
*/
sanitize_error_detail: boolean | null;
/**
* Scan Only Tool Results
* @description When True, unified guardrails only evaluate tool results, the untrusted data an agent feeds back into the model, and skip system, user, and assistant content. Intended for agent harnesses whose own prompt scaffolding is trusted but often trips prompt-attack detectors.
*/
scan_only_tool_results?: boolean | null;
/**
* Scan Raw Request
* @description When True, this pre_call guardrail always evaluates the request as it was before any guardrail in this hook ran, regardless of its position in the guardrails list -- so the YAML order of guardrails can never change whether this one blocks. Use only for block-only guardrails: any data this guardrail returns is discarded, same contract as run_in_parallel, since an earlier guardrail's masking must not be undone by this one.
*/
scan_raw_request?: boolean | null;
/**
* Sensitive Data Route To Model
* @description Model to route requests to when sensitive data is detected and on_sensitive_data='route'. This is typically an on-premise model for data privacy. The routing decision persists for the entire session.
*/
sensitive_data_route_to_model?: string | null;
/**
* Severity Threshold
* @description Minimum severity to block (high, medium, low)
*/
severity_threshold?: string | null;
/**
* Skip System Message In Guardrail
* @description When True, unified guardrails skip system-role messages when building evaluation inputs (texts and structured_messages). When False, system messages are included even if litellm_settings sets a global skip. When None, use the global litellm.skip_system_message_in_guardrail setting. For Anthropic /v1/messages, the flag applies only to the trusted top-level system prompt. In-sequence system entries are untrusted client input and remain in texts and structured_messages.
*/
skip_system_message_in_guardrail?: boolean | null;
/**
* Skip Tool Message In Guardrail
* @description When True, unified guardrails skip tool-role messages when building evaluation inputs (texts and structured_messages). When False, tool messages are included even if litellm_settings sets a global skip. When None, use the global litellm.skip_tool_message_in_guardrail setting.
*/
skip_tool_message_in_guardrail?: boolean | null;
/**
* Skip Unscannable Attachments
* @description Implemented by guardrail='model_armor' and guardrail='bedrock'. When True, attachments the guardrail cannot scan pass through unscanned instead of blocking. For Model Armor these are references with no inline bytes (file_id, gs://, or http(s) URLs), and fail_on_error still governs real Model Armor API errors. For Bedrock these are documents, files, audio, video, and images that are not inline PNG or JPEG up to 4 MB. Default False blocks them.
* @default false
*/
skip_unscannable_attachments: boolean | null;
/**
* Sticky Session Routing
* @description When True (default), after sensitive data is detected and routed, all subsequent requests in the same session will continue routing to the same model.
* @default true
*/
sticky_session_routing: boolean | null;
/**
* Stream Scope
* @description Whether this guardrail runs on streaming requests, non-streaming requests, or both. A string applies to every configured mode. A map overrides named modes (pre_call, during_call, post_call, ...); omitted keys default to both. Unset means both, matching historical behavior.
*/
stream_scope?: ("streaming" | "non_streaming" | "both") | {
[key: string]: "streaming" | "non_streaming" | "both";
} | null;
/**
* Template Id
* @description The ID of your Model Armor template
*/
template_id?: string | null;
/**
* Timeout
* @description Per-request timeout for the guardrail provider API call (seconds). Accepts int, float, or numeric string; coerced to float on load. Each guardrail handler chooses its own default when unset.
*/
timeout?: number | null;
/**
* Unreachable Fallback
* @description Behavior when a guardrail endpoint is unreachable due to network errors. Implemented by guardrail='generic_guardrail_api', 'agent_365', 'akto', 'vigil_guard', 'repelloai', 'headroom', 'compresr', and 'typesafe'. 'fail_closed' raises an error (default). 'fail_open' logs a critical error and allows the request to proceed. Also implemented by guardrail='decision_model'.
* @default fail_closed
* @enum {string}
*/
unreachable_fallback: "fail_closed" | "fail_open";
/**
* Violation Message Template
* @description Custom message when a guardrail blocks an action. Supports placeholders like {tool_name}, {rule_id}, and {default_message}.
*/
violation_message_template?: string | null;
} & {
[key: string]: unknown;
};
/** BaseModel */
BaseModel: Record<string, never>;
/**
* BedrockChecksConfigModel
* @description Inline `checks` config for the resource-less Bedrock InvokeGuardrailChecks API.
*
* Include only the checks you want to run; at least one must be set.
*/
BedrockChecksConfigModel: {
contentFilter?: components["schemas"]["BedrockChecksContentFilterModel"] | null;
promptAttack?: components["schemas"]["BedrockChecksPromptAttackModel"] | null;
sensitiveInformation?: components["schemas"]["BedrockChecksSensitiveInformationModel"] | null;
};
/** BedrockChecksContentFilterCategoryItem */
BedrockChecksContentFilterCategoryItem: {
/**
* Category
* @enum {string}
*/
category: "VIOLENCE" | "HATE" | "SEXUAL" | "MISCONDUCT" | "INSULTS";
};
/** BedrockChecksContentFilterModel */
BedrockChecksContentFilterModel: {
/** Categories */
categories: components["schemas"]["BedrockChecksContentFilterCategoryItem"][];
};
/** BedrockChecksPromptAttackCategoryItem */
BedrockChecksPromptAttackCategoryItem: {
/**
* Category
* @enum {string}
*/
category: "JAILBREAK" | "PROMPT_INJECTION" | "PROMPT_LEAKAGE";
};
/** BedrockChecksPromptAttackModel */
BedrockChecksPromptAttackModel: {
/** Categories */
categories: components["schemas"]["BedrockChecksPromptAttackCategoryItem"][];
};
/** BedrockChecksSensitiveInformationEntityItem */
BedrockChecksSensitiveInformationEntityItem: {
/**
* Type
* @enum {string}
*/
type: "ADDRESS" | "AGE" | "AWS_ACCESS_KEY" | "AWS_SECRET_KEY" | "CA_HEALTH_NUMBER" | "CA_SOCIAL_INSURANCE_NUMBER" | "CREDIT_DEBIT_CARD_CVV" | "CREDIT_DEBIT_CARD_EXPIRY" | "CREDIT_DEBIT_CARD_NUMBER" | "DRIVER_ID" | "EMAIL" | "INTERNATIONAL_BANK_ACCOUNT_NUMBER" | "IP_ADDRESS" | "LICENSE_PLATE" | "MAC_ADDRESS" | "NAME" | "PASSWORD" | "PHONE" | "PIN" | "SWIFT_CODE" | "UK_NATIONAL_HEALTH_SERVICE_NUMBER" | "UK_NATIONAL_INSURANCE_NUMBER" | "UK_UNIQUE_TAXPAYER_REFERENCE_NUMBER" | "URL" | "USERNAME" | "US_BANK_ACCOUNT_NUMBER" | "US_BANK_ROUTING_NUMBER" | "US_INDIVIDUAL_TAX_IDENTIFICATION_NUMBER" | "US_PASSPORT_NUMBER" | "US_SOCIAL_SECURITY_NUMBER" | "VEHICLE_IDENTIFICATION_NUMBER";
};
/** BedrockChecksSensitiveInformationModel */
BedrockChecksSensitiveInformationModel: {
/** Entities */
entities: components["schemas"]["BedrockChecksSensitiveInformationEntityItem"][];
};
/** BlockKeyRequest */
BlockKeyRequest: {
/** Key */
key: string;
};
/** BlockModelRequest */
BlockModelRequest: {
/** Model Id */
model_id: string;
};
/** BlockTeamRequest */
BlockTeamRequest: {
/** Team Id */
team_id: string;
};
/** BlockUnpricedModelsRequest */
BlockUnpricedModelsRequest: {
/** Enabled */
enabled: boolean;
};
/** BlockUnpricedModelsResponse */
BlockUnpricedModelsResponse: {
/** Enabled */
enabled: boolean;
};
/** BlockUsers */
BlockUsers: {
/** User Ids */
user_ids: string[];
};
/** BlockUsersResponse */
BlockUsersResponse: {
/** Blocked Users */
blocked_users: components["schemas"]["LiteLLM_EndUserTable"][];
};
/**
* BlockedWord
* @description Represents a blocked word with its action and optional description
*/
BlockedWord: {
/** @description Action to take when keyword is detected (BLOCK or MASK) */
action: components["schemas"]["ContentFilterAction"];
/**
* Description
* @description Optional description explaining why this keyword is sensitive
*/
description?: string | null;
/**
* Keyword
* @description The keyword to block or mask
*/
keyword: string;
};
/** BlogPost */
BlogPost: {
/** Date */
date: string;
/** Description */
description: string;
/**
* Tags
* @default []
*/
tags: string[];
/** Title */
title: string;
/** Url */
url: string;
};
/** BlogPostsResponse */
BlogPostsResponse: {
/** Posts */
posts: components["schemas"]["BlogPost"][];
};
/** Body_audio_transcriptions_audio_transcriptions_post */
Body_audio_transcriptions_audio_transcriptions_post: {
/** File */
file: string;
};
/** Body_audio_transcriptions_v1_audio_transcriptions_post */
Body_audio_transcriptions_v1_audio_transcriptions_post: {
/** File */
file: string;
};
/** Body_authorize_complete_authorize_complete_post */
Body_authorize_complete_authorize_complete_post: {
/** Decision */
decision?: string | null;
/** Delivery */
delivery?: string | null;
/** Flow */
flow: string;
/** Team Id */
team_id?: string | null;
};
/** Body_convert_prompt_file_to_json_utils_dotprompt_json_converter_post */
Body_convert_prompt_file_to_json_utils_dotprompt_json_converter_post: {
/** File */
file: string;
};
/** Body_create_file__provider__v1_files_post */
Body_create_file__provider__v1_files_post: {
/**
* Custom Llm Provider
* @default openai
*/
custom_llm_provider: string;
/** File */
file: string;
/** Litellm Metadata */
litellm_metadata?: string | null;
/**
* Passthrough
* @default false
*/
passthrough: boolean;
/** Purpose */
purpose: string;
/**
* Target Model Names
* @default
*/
target_model_names: string;
/**
* Target Storage
* @default default
*/
target_storage: string;
};
/** Body_create_file_files_post */
Body_create_file_files_post: {
/**
* Custom Llm Provider
* @default openai
*/
custom_llm_provider: string;
/** File */
file: string;
/** Litellm Metadata */
litellm_metadata?: string | null;
/**
* Passthrough
* @default false
*/
passthrough: boolean;
/** Purpose */
purpose: string;
/**
* Target Model Names
* @default
*/
target_model_names: string;
/**
* Target StFile truncated at 1 MB.