A self-improving RLM agent for coding workflows and long-running autonomous tasks.
Rust
//! Compiled fallback catalog ported from the TS `packages/ai/src/models.generated.ts`.
//!
//! The full catalog is fetched at runtime (no-cold-start chain: validated disk cache ->
//! packaged bundled snapshot -> this compiled fallback). Hand-maintained; there is no generator.
use std::collections::HashMap;
use std::sync::LazyLock;
use crate::types::{
CompatKind, JsNumber, MaxTokensField, Model, ModelCompat, ModelCost, ModelInput,
ModelThinkingLevel, OpenAiCompletionsCompat, ThinkingFormat,
};
type ProviderCatalog = HashMap<String, HashMap<String, Model>>;
/// The compat fields the fallback table carries: the
/// [`OpenAiCompletionsCompat`] subset the TS fallback file uses. Absent
/// fields stay `None` on the wire.
struct RawCompat {
supports_store: Option<bool>,
supports_developer_role: Option<bool>,
supports_reasoning_effort: Option<bool>,
max_tokens_field: Option<MaxTokensField>,
requires_reasoning_content_on_assistant_messages: Option<bool>,
thinking_format: Option<ThinkingFormat>,
zai_tool_stream: Option<bool>,
send_session_affinity_headers: Option<bool>,
supports_strict_mode: Option<bool>,
}
/// One entry of the compiled fallback table, in the TS `MODELS` layout.
struct RawModel {
provider: &'static str,
id: &'static str,
name: &'static str,
api: &'static str,
base_url: &'static str,
reasoning: bool,
/// The `thinkingLevelMap` pairs, in the TS level order; `None` marks a
/// level as unsupported, an absent level is simply not listed.
thinking_levels: &'static [(ModelThinkingLevel, Option<&'static str>)],
input: &'static [ModelInput],
/// (`input`, `output`, `cacheRead`, `cacheWrite`) in $/million tokens.
cost: (f64, f64, f64, f64),
context_window: u64,
max_tokens: u64,
featured: Option<bool>,
headers: &'static [(&'static str, &'static str)],
compat: Option<RawCompat>,
}
static RAW_MODELS: &[RawModel] = &[
RawModel {
provider: "amazon-bedrock",
id: "amazon.nova-2-lite-v1:0",
name: "Nova 2 Lite",
api: "bedrock-converse-stream",
base_url: "https://bedrock-runtime.us-east-1.amazonaws.com",
reasoning: true,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (0.33, 2.75, 0.0, 0.0),
context_window: 128_000,
max_tokens: 4_096,
featured: None,
headers: &[],
compat: None,
},
RawModel {
provider: "amazon-bedrock",
id: "au.anthropic.claude-haiku-4-5-20251001-v1:0",
name: "Claude Haiku 4.5 (AU)",
api: "bedrock-converse-stream",
base_url: "https://bedrock-runtime.ap-southeast-2.amazonaws.com",
reasoning: true,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (1.0, 5.0, 0.1, 1.25),
context_window: 200_000,
max_tokens: 64_000,
featured: None,
headers: &[],
compat: None,
},
RawModel {
provider: "amazon-bedrock",
id: "eu.anthropic.claude-fable-5",
name: "Claude Fable 5 (EU)",
api: "bedrock-converse-stream",
base_url: "https://bedrock-runtime.eu-central-1.amazonaws.com",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Off, None),
(crate::types::ModelThinkingLevel::Xhigh, Some("xhigh")),
(crate::types::ModelThinkingLevel::Max, Some("max")),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (11.0, 55.0, 1.1, 13.75),
context_window: 1_000_000,
max_tokens: 128_000,
featured: None,
headers: &[],
compat: None,
},
RawModel {
provider: "amazon-bedrock",
id: "jp.anthropic.claude-haiku-4-5-20251001-v1:0",
name: "Claude Haiku 4.5 (JP)",
api: "bedrock-converse-stream",
base_url: "https://bedrock-runtime.ap-northeast-1.amazonaws.com",
reasoning: true,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (1.0, 5.0, 0.1, 1.25),
context_window: 200_000,
max_tokens: 64_000,
featured: None,
headers: &[],
compat: None,
},
RawModel {
provider: "anthropic",
id: "claude-fable-5",
name: "Claude Fable 5",
api: "anthropic-messages",
base_url: "https://api.anthropic.com",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Off, None),
(crate::types::ModelThinkingLevel::Xhigh, Some("xhigh")),
(crate::types::ModelThinkingLevel::Max, Some("max")),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (10.0, 50.0, 1.0, 12.5),
context_window: 1_000_000,
max_tokens: 128_000,
featured: None,
headers: &[],
compat: None,
},
RawModel {
provider: "azure-openai-responses",
id: "gpt-4",
name: "GPT-4",
api: "azure-openai-responses",
base_url: "",
reasoning: false,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text],
cost: (30.0, 60.0, 0.0, 0.0),
context_window: 8_192,
max_tokens: 8_192,
featured: None,
headers: &[],
compat: None,
},
RawModel {
provider: "cerebras",
id: "gemma-4-31b",
name: "Gemma 4 31B IT",
api: "openai-completions",
base_url: "https://api.cerebras.ai/v1",
reasoning: true,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (0.99, 1.49, 0.0, 0.0),
context_window: 131_072,
max_tokens: 40_960,
featured: None,
headers: &[],
compat: None,
},
RawModel {
provider: "cloudflare-ai-gateway",
id: "claude-fable-5",
name: "Claude Fable 5",
api: "anthropic-messages",
base_url: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/anthropic",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Off, None),
(crate::types::ModelThinkingLevel::Xhigh, Some("xhigh")),
(crate::types::ModelThinkingLevel::Max, Some("max")),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (10.0, 50.0, 1.0, 12.5),
context_window: 1_000_000,
max_tokens: 128_000,
featured: None,
headers: &[],
compat: None,
},
RawModel {
provider: "cloudflare-ai-gateway",
id: "gpt-4.1",
name: "GPT-4.1",
api: "openai-responses",
base_url: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/openai",
reasoning: false,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (2.0, 8.0, 0.5, 0.0),
context_window: 1_047_576,
max_tokens: 32_768,
featured: None,
headers: &[],
compat: None,
},
RawModel {
provider: "cloudflare-workers-ai",
id: "@cf/deepseek-ai/deepseek-v4-flash-0731",
name: "DeepSeek V4 Flash 0731",
api: "openai-completions",
base_url: "https://api.cloudflare.com/client/v4/accounts/{CLOUDFLARE_ACCOUNT_ID}/ai/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, None),
(crate::types::ModelThinkingLevel::Medium, None),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, Some("max")),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text],
cost: (0.44, 1.32, 0.014, 0.0),
context_window: 1_310_720,
max_tokens: 1_048_576,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: None,
supports_developer_role: None,
supports_reasoning_effort: None,
max_tokens_field: None,
requires_reasoning_content_on_assistant_messages: Some(true),
thinking_format: Some(crate::types::ThinkingFormat::Deepseek),
zai_tool_stream: None,
send_session_affinity_headers: Some(true),
supports_strict_mode: None,
}),
},
RawModel {
provider: "deepseek",
id: "deepseek-v4-flash",
name: "DeepSeek V4 Flash",
api: "openai-completions",
base_url: "https://api.deepseek.com",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, None),
(crate::types::ModelThinkingLevel::Medium, None),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, Some("max")),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text],
cost: (0.14, 0.28, 0.002_8, 0.0),
context_window: 1_000_000,
max_tokens: 384_000,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: None,
supports_developer_role: None,
supports_reasoning_effort: None,
max_tokens_field: None,
requires_reasoning_content_on_assistant_messages: Some(true),
thinking_format: Some(crate::types::ThinkingFormat::Deepseek),
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: None,
}),
},
RawModel {
provider: "fireworks",
id: "accounts/fireworks/models/deepseek-v4-flash-0731",
name: "DeepSeek V4 Flash 0731",
api: "anthropic-messages",
base_url: "https://api.fireworks.ai/inference",
reasoning: true,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text],
cost: (0.22, 0.66, 0.007, 0.0),
context_window: 1_000_000,
max_tokens: 384_000,
featured: None,
headers: &[],
compat: None,
},
RawModel {
provider: "github-copilot",
id: "claude-fable-5",
name: "Claude Fable 5",
api: "anthropic-messages",
base_url: "https://api.individual.githubcopilot.com",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Off, None),
(crate::types::ModelThinkingLevel::Xhigh, Some("xhigh")),
(crate::types::ModelThinkingLevel::Max, Some("max")),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (10.0, 50.0, 1.0, 12.5),
context_window: 1_000_000,
max_tokens: 128_000,
featured: None,
headers: &[
("User-Agent", "GitHubCopilotChat/0.48.1"),
("Editor-Version", "vscode/1.136.1"),
("Editor-Plugin-Version", "copilot-chat/0.48.1"),
("Copilot-Integration-Id", "vscode-chat"),
],
compat: None,
},
RawModel {
provider: "github-copilot",
id: "gemini-3.5-flash",
name: "Gemini 3.5 Flash",
api: "openai-completions",
base_url: "https://api.individual.githubcopilot.com",
reasoning: true,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (1.5, 9.0, 0.15, 0.0),
context_window: 200_000,
max_tokens: 64_000,
featured: None,
headers: &[
("User-Agent", "GitHubCopilotChat/0.48.1"),
("Editor-Version", "vscode/1.136.1"),
("Editor-Plugin-Version", "copilot-chat/0.48.1"),
("Copilot-Integration-Id", "vscode-chat"),
],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(false),
max_tokens_field: None,
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: None,
}),
},
RawModel {
provider: "github-copilot",
id: "gpt-5-mini",
name: "GPT-5 Mini",
api: "openai-responses",
base_url: "https://api.individual.githubcopilot.com",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Off, None),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (0.25, 2.0, 0.025, 0.0),
context_window: 264_000,
max_tokens: 64_000,
featured: None,
headers: &[
("User-Agent", "GitHubCopilotChat/0.48.1"),
("Editor-Version", "vscode/1.136.1"),
("Editor-Plugin-Version", "copilot-chat/0.48.1"),
("Copilot-Integration-Id", "vscode-chat"),
],
compat: None,
},
RawModel {
provider: "google",
id: "gemini-2.5-flash",
name: "Gemini 2.5 Flash",
api: "google-generative-ai",
base_url: "https://generativelanguage.googleapis.com/v1beta",
reasoning: true,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (0.3, 2.5, 0.03, 0.0),
context_window: 1_048_576,
max_tokens: 65_536,
featured: None,
headers: &[],
compat: None,
},
RawModel {
provider: "google-vertex",
id: "gemini-1.5-flash",
name: "Gemini 1.5 Flash (Vertex)",
api: "google-vertex",
base_url: "https://{location}-aiplatform.googleapis.com",
reasoning: false,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (0.075, 0.3, 0.018_75, 0.0),
context_window: 1_000_000,
max_tokens: 8_192,
featured: None,
headers: &[],
compat: None,
},
RawModel {
provider: "groq",
id: "llama-3.1-8b-instant",
name: "Llama 3.1 8B",
api: "openai-completions",
base_url: "https://api.groq.com/openai/v1",
reasoning: false,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text],
cost: (0.05, 0.08, 0.0, 0.0),
context_window: 131_072,
max_tokens: 131_072,
featured: None,
headers: &[],
compat: None,
},
RawModel {
provider: "huggingface",
id: "deepseek-ai/DeepSeek-R1",
name: "DeepSeek-R1",
api: "openai-completions",
base_url: "https://router.huggingface.co/v1",
reasoning: true,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text],
cost: (0.7, 2.5, 0.0, 0.0),
context_window: 64_000,
max_tokens: 32_768,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: None,
supports_developer_role: Some(false),
supports_reasoning_effort: None,
max_tokens_field: None,
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: None,
}),
},
RawModel {
provider: "kimi-coding",
id: "k3",
name: "Kimi K3",
api: "anthropic-messages",
base_url: "https://api.kimi.com/coding",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Off, None),
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, None),
(crate::types::ModelThinkingLevel::Medium, None),
(crate::types::ModelThinkingLevel::High, None),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, Some("max")),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (0.0, 0.0, 0.0, 0.0),
context_window: 1_048_576,
max_tokens: 131_072,
featured: None,
headers: &[
("User-Agent", "KimiCLI/1.5"),
],
compat: None,
},
RawModel {
provider: "minimax",
id: "MiniMax-M2.7",
name: "MiniMax-M2.7",
api: "anthropic-messages",
base_url: "https://api.minimax.io/anthropic",
reasoning: true,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text],
cost: (0.3, 1.2, 0.06, 0.375),
context_window: 204_800,
max_tokens: 131_072,
featured: None,
headers: &[],
compat: None,
},
RawModel {
provider: "minimax-cn",
id: "MiniMax-M2.7",
name: "MiniMax-M2.7",
api: "anthropic-messages",
base_url: "https://api.minimaxi.com/anthropic",
reasoning: true,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text],
cost: (0.3, 1.2, 0.06, 0.375),
context_window: 204_800,
max_tokens: 131_072,
featured: None,
headers: &[],
compat: None,
},
RawModel {
provider: "mistral",
id: "codestral-latest",
name: "Codestral (latest)",
api: "mistral-conversations",
base_url: "https://api.mistral.ai",
reasoning: false,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text],
cost: (0.3, 0.9, 0.0, 0.0),
context_window: 256_000,
max_tokens: 4_096,
featured: None,
headers: &[],
compat: None,
},
RawModel {
provider: "moonshotai",
id: "kimi-k2-0711-preview",
name: "Kimi K2 0711",
api: "openai-completions",
base_url: "https://api.moonshot.ai/v1",
reasoning: false,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text],
cost: (0.6, 2.5, 0.15, 0.0),
context_window: 131_072,
max_tokens: 16_384,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(false),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "moonshotai-cn",
id: "kimi-k2-0711-preview",
name: "Kimi K2 0711",
api: "openai-completions",
base_url: "https://api.moonshot.cn/v1",
reasoning: false,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text],
cost: (0.6, 2.5, 0.15, 0.0),
context_window: 131_072,
max_tokens: 16_384,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(false),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "openai",
id: "gpt-4",
name: "GPT-4",
api: "openai-responses",
base_url: "https://api.openai.com/v1",
reasoning: false,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text],
cost: (30.0, 60.0, 0.0, 0.0),
context_window: 8_192,
max_tokens: 8_192,
featured: None,
headers: &[],
compat: None,
},
RawModel {
provider: "openai-codex",
id: "gpt-5.1",
name: "GPT-5.1",
api: "openai-codex-responses",
base_url: "https://chatgpt.com/backend-api",
reasoning: true,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (1.25, 10.0, 0.125, 0.0),
context_window: 272_000,
max_tokens: 128_000,
featured: None,
headers: &[],
compat: None,
},
RawModel {
provider: "opencode",
id: "big-pickle",
name: "Big Pickle",
api: "openai-completions",
base_url: "https://opencode.ai/zen/v1",
reasoning: true,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text],
cost: (0.0, 0.0, 0.0, 0.0),
context_window: 200_000,
max_tokens: 32_000,
featured: None,
headers: &[],
compat: None,
},
RawModel {
provider: "opencode",
id: "claude-fable-5",
name: "Claude Fable 5",
api: "anthropic-messages",
base_url: "https://opencode.ai/zen",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Off, None),
(crate::types::ModelThinkingLevel::Xhigh, Some("xhigh")),
(crate::types::ModelThinkingLevel::Max, Some("max")),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (10.0, 50.0, 1.0, 12.5),
context_window: 1_000_000,
max_tokens: 128_000,
featured: None,
headers: &[],
compat: None,
},
RawModel {
provider: "opencode",
id: "gemini-3-flash",
name: "Gemini 3 Flash",
api: "google-generative-ai",
base_url: "https://opencode.ai/zen/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Off, None),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (0.5, 3.0, 0.05, 0.0),
context_window: 1_048_576,
max_tokens: 65_536,
featured: None,
headers: &[],
compat: None,
},
RawModel {
provider: "opencode",
id: "gpt-5",
name: "GPT-5",
api: "openai-responses",
base_url: "https://opencode.ai/zen/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Off, None),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (1.07, 8.5, 0.107, 0.0),
context_window: 400_000,
max_tokens: 128_000,
featured: None,
headers: &[],
compat: None,
},
RawModel {
provider: "opencode-go",
id: "deepseek-v4-flash",
name: "DeepSeek V4 Flash",
api: "openai-completions",
base_url: "https://opencode.ai/zen/go/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, None),
(crate::types::ModelThinkingLevel::Medium, None),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, Some("max")),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text],
cost: (0.22, 0.66, 0.007, 0.0),
context_window: 1_000_000,
max_tokens: 384_000,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: None,
supports_developer_role: None,
supports_reasoning_effort: None,
max_tokens_field: None,
requires_reasoning_content_on_assistant_messages: Some(true),
thinking_format: Some(crate::types::ThinkingFormat::Deepseek),
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: None,
}),
},
RawModel {
provider: "opencode-go",
id: "gpt-5.6-luna",
name: "GPT-5.6 Luna",
api: "openai-responses",
base_url: "https://opencode.ai/zen/go/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Off, None),
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Xhigh, Some("xhigh")),
(crate::types::ModelThinkingLevel::Max, Some("max")),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (0.2, 1.2, 0.02, 0.25),
context_window: 1_050_000,
max_tokens: 128_000,
featured: None,
headers: &[],
compat: None,
},
RawModel {
provider: "opencode-go",
id: "minimax-m3",
name: "MiniMax-M3",
api: "anthropic-messages",
base_url: "https://opencode.ai/zen/go",
reasoning: true,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (0.3, 1.2, 0.06, 0.0),
context_window: 1_000_000,
max_tokens: 131_072,
featured: None,
headers: &[],
compat: None,
},
RawModel {
provider: "openrouter",
id: "~anthropic/claude-fable-latest",
name: "Anthropic: Claude Fable Latest",
api: "openai-completions",
base_url: "https://openrouter.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Off, None),
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, Some("low")),
(crate::types::ModelThinkingLevel::Medium, Some("medium")),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, Some("xhigh")),
(crate::types::ModelThinkingLevel::Max, Some("max")),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (10.0, 50.0, 0.25, 12.5),
context_window: 1_000_000,
max_tokens: 128_000,
featured: None,
headers: &[],
compat: None,
},
RawModel {
provider: "prime-inference",
id: "anthropic/claude-fable-5",
name: "Claude Fable 5",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Off, None),
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, Some("low")),
(crate::types::ModelThinkingLevel::Medium, Some("medium")),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, Some("xhigh")),
(crate::types::ModelThinkingLevel::Max, Some("max")),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (10.0, 50.0, 1.0, 12.5),
context_window: 1_000_000,
max_tokens: 128_000,
featured: Some(true),
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "anthropic/claude-fable-5.1",
name: "Claude Fable 5.1",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Off, None),
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, Some("low")),
(crate::types::ModelThinkingLevel::Medium, Some("medium")),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, Some("xhigh")),
(crate::types::ModelThinkingLevel::Max, Some("max")),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (10.0, 50.0, 1.0, 12.5),
context_window: 1_000_000,
max_tokens: 128_000,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "anthropic/claude-haiku-4.5",
name: "Claude Haiku 4.5",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, None),
(crate::types::ModelThinkingLevel::Medium, None),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (1.0, 5.0, 0.1, 1.25),
context_window: 200_000,
max_tokens: 64_000,
featured: Some(true),
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(false),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: Some(crate::types::ThinkingFormat::Openrouter),
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "anthropic/claude-opus-4.1",
name: "Claude Opus 4.1",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, None),
(crate::types::ModelThinkingLevel::Medium, None),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (15.0, 75.0, 1.5, 18.75),
context_window: 200_000,
max_tokens: 32_000,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(false),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: Some(crate::types::ThinkingFormat::Openrouter),
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "anthropic/claude-opus-4.5",
name: "Claude Opus 4.5",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, None),
(crate::types::ModelThinkingLevel::Medium, None),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (5.0, 25.0, 0.5, 6.25),
context_window: 200_000,
max_tokens: 64_000,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(false),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: Some(crate::types::ThinkingFormat::Openrouter),
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "anthropic/claude-opus-4.6",
name: "Claude Opus 4.6",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, Some("low")),
(crate::types::ModelThinkingLevel::Medium, Some("medium")),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, Some("max")),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (5.0, 25.0, 0.5, 6.25),
context_window: 1_000_000,
max_tokens: 128_000,
featured: Some(true),
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "anthropic/claude-opus-4.7",
name: "Claude Opus 4.7",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, Some("low")),
(crate::types::ModelThinkingLevel::Medium, Some("medium")),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, Some("xhigh")),
(crate::types::ModelThinkingLevel::Max, Some("max")),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (5.0, 25.0, 0.5, 6.25),
context_window: 1_000_000,
max_tokens: 128_000,
featured: Some(true),
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "anthropic/claude-opus-4.8",
name: "Claude Opus 4.8",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, Some("low")),
(crate::types::ModelThinkingLevel::Medium, Some("medium")),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, Some("xhigh")),
(crate::types::ModelThinkingLevel::Max, Some("max")),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (5.0, 25.0, 0.5, 6.25),
context_window: 1_000_000,
max_tokens: 128_000,
featured: Some(true),
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "anthropic/claude-opus-5",
name: "Claude Opus 5",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, Some("low")),
(crate::types::ModelThinkingLevel::Medium, Some("medium")),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, Some("xhigh")),
(crate::types::ModelThinkingLevel::Max, Some("max")),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (5.0, 25.0, 0.5, 6.25),
context_window: 1_000_000,
max_tokens: 128_000,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "anthropic/claude-sonnet-4",
name: "Claude Sonnet 4",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, None),
(crate::types::ModelThinkingLevel::Medium, None),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (3.0, 15.0, 0.300_000_000_000_000_04, 3.75),
context_window: 200_000,
max_tokens: 64_000,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(false),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: Some(crate::types::ThinkingFormat::Openrouter),
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "anthropic/claude-sonnet-4.5",
name: "Claude Sonnet 4.5",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, None),
(crate::types::ModelThinkingLevel::Medium, None),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (3.0, 15.0, 0.300_000_000_000_000_04, 3.75),
context_window: 200_000,
max_tokens: 64_000,
featured: Some(true),
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(false),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: Some(crate::types::ThinkingFormat::Openrouter),
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "anthropic/claude-sonnet-4.6",
name: "Claude Sonnet 4.6",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, Some("low")),
(crate::types::ModelThinkingLevel::Medium, Some("medium")),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, Some("max")),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (3.0, 15.0, 0.300_000_000_000_000_04, 3.75),
context_window: 1_000_000,
max_tokens: 128_000,
featured: Some(true),
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "anthropic/claude-sonnet-5",
name: "Claude Sonnet 5",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, Some("low")),
(crate::types::ModelThinkingLevel::Medium, Some("medium")),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, Some("xhigh")),
(crate::types::ModelThinkingLevel::Max, Some("max")),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (2.0, 10.0, 0.2, 2.5),
context_window: 1_000_000,
max_tokens: 128_000,
featured: Some(true),
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "deepseek/deepseek-chat",
name: "Deepseek Chat",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: false,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text],
cost: (0.4, 1.3, 0.0, 0.0),
context_window: 163_840,
max_tokens: 16_384,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "deepseek/deepseek-chat-v3-0324",
name: "Deepseek Chat V3 0324",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: false,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text],
cost: (0.27, 1.12, 0.0, 0.0),
context_window: 163_840,
max_tokens: 147_456,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "deepseek/deepseek-chat-v3.1",
name: "Deepseek Chat V3.1",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, None),
(crate::types::ModelThinkingLevel::Medium, None),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text],
cost: (0.6, 1.7, 0.0, 0.0),
context_window: 163_840,
max_tokens: 144_900,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(false),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: Some(crate::types::ThinkingFormat::Openrouter),
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "deepseek/deepseek-v3.1-terminus",
name: "Deepseek V3.1 Terminus",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, None),
(crate::types::ModelThinkingLevel::Medium, None),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text],
cost: (0.27, 1.0, 0.0, 0.0),
context_window: 163_840,
max_tokens: 32_768,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(false),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: Some(crate::types::ThinkingFormat::Openrouter),
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "deepseek/deepseek-v3.2",
name: "Deepseek V3.2",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, None),
(crate::types::ModelThinkingLevel::Medium, None),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text],
cost: (0.370_5, 1.111_5, 0.0, 0.0),
context_window: 163_840,
max_tokens: 65_536,
featured: Some(true),
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(false),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: Some(crate::types::ThinkingFormat::Openrouter),
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "deepseek/deepseek-v3.2-exp",
name: "Deepseek V3.2 EXP",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, None),
(crate::types::ModelThinkingLevel::Medium, None),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text],
cost: (0.27, 0.41, 0.0, 0.0),
context_window: 163_840,
max_tokens: 65_536,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(false),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: Some(crate::types::ThinkingFormat::Openrouter),
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "deepseek/deepseek-v4-flash",
name: "Deepseek V4 Flash",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, None),
(crate::types::ModelThinkingLevel::Medium, None),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, Some("max")),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text],
cost: (0.44, 1.32, 0.0, 0.0),
context_window: 1_048_576,
max_tokens: 384_000,
featured: Some(true),
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: Some(true),
thinking_format: Some(crate::types::ThinkingFormat::Deepseek),
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "deepseek/deepseek-v4-flash-0731",
name: "Deepseek V4 Flash 0731",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, None),
(crate::types::ModelThinkingLevel::Medium, None),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, Some("max")),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text],
cost: (0.44, 1.32, 0.0, 0.0),
context_window: 1_310_720,
max_tokens: 131_072,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: Some(true),
thinking_format: Some(crate::types::ThinkingFormat::Deepseek),
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "deepseek/deepseek-v4-pro",
name: "Deepseek V4 PRO",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, None),
(crate::types::ModelThinkingLevel::Medium, None),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, Some("max")),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text],
cost: (1.91, 3.83, 0.0, 0.0),
context_window: 1_048_576,
max_tokens: 384_000,
featured: Some(true),
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: Some(true),
thinking_format: Some(crate::types::ThinkingFormat::Deepseek),
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "google/gemini-2.5-flash",
name: "Gemini 2.5 Flash",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, None),
(crate::types::ModelThinkingLevel::Medium, None),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (0.3, 2.5, 0.0, 0.0),
context_window: 1_048_576,
max_tokens: 65_535,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(false),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: Some(crate::types::ThinkingFormat::Openrouter),
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "google/gemini-2.5-flash-lite",
name: "Gemini 2.5 Flash Lite",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, None),
(crate::types::ModelThinkingLevel::Medium, None),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (0.1, 0.4, 0.0, 0.0),
context_window: 1_048_576,
max_tokens: 65_535,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(false),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: Some(crate::types::ThinkingFormat::Openrouter),
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "google/gemini-2.5-pro",
name: "Gemini 2.5 PRO",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Off, None),
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, None),
(crate::types::ModelThinkingLevel::Medium, None),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (1.25, 10.0, 0.0, 0.0),
context_window: 1_048_576,
max_tokens: 65_536,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(false),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: Some(crate::types::ThinkingFormat::Openrouter),
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "google/gemini-3-flash-preview",
name: "Gemini 3 Flash Preview",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, Some("minimal")),
(crate::types::ModelThinkingLevel::Low, Some("low")),
(crate::types::ModelThinkingLevel::Medium, Some("medium")),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (0.5, 3.0, 0.0, 0.0),
context_window: 1_048_576,
max_tokens: 65_536,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "google/gemini-3.1-pro-preview",
name: "Gemini 3.1 PRO Preview",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Off, None),
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, Some("low")),
(crate::types::ModelThinkingLevel::Medium, Some("medium")),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (2.0, 12.0, 0.0, 0.0),
context_window: 1_048_576,
max_tokens: 65_536,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "google/gemini-3.5-flash",
name: "Gemini 3.5 Flash",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Off, None),
(crate::types::ModelThinkingLevel::Minimal, Some("minimal")),
(crate::types::ModelThinkingLevel::Low, Some("low")),
(crate::types::ModelThinkingLevel::Medium, Some("medium")),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (1.5, 9.0, 0.0, 0.0),
context_window: 1_048_576,
max_tokens: 65_536,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "google/gemini-3.6-flash",
name: "Gemini 3.6 Flash",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Off, None),
(crate::types::ModelThinkingLevel::Minimal, Some("minimal")),
(crate::types::ModelThinkingLevel::Low, Some("low")),
(crate::types::ModelThinkingLevel::Medium, Some("medium")),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (1.5, 7.5, 0.0, 0.0),
context_window: 1_048_576,
max_tokens: 65_536,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "google/gemini-3.7-flash",
name: "Gemini 3.7 Flash",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Off, None),
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, Some("low")),
(crate::types::ModelThinkingLevel::Medium, Some("medium")),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (1.35, 6.75, 0.0, 0.0),
context_window: 1_048_576,
max_tokens: 65_536,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "google/gemini-3.8-flash",
name: "Gemini 3.8 Flash",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Off, None),
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, Some("low")),
(crate::types::ModelThinkingLevel::Medium, Some("medium")),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (0.75, 3.75, 0.0, 0.0),
context_window: 1_048_576,
max_tokens: 65_536,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "google/gemma-3-27b-it",
name: "Gemma 3 27B IT",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: false,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (0.119, 0.3, 0.0, 0.0),
context_window: 131_072,
max_tokens: 117_964,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "meta-llama/Llama-3.2-1B-Instruct",
name: "Llama 3.2 1B Instruct",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: false,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text],
cost: (0.027, 0.201, 0.0, 0.0),
context_window: 60_000,
max_tokens: 54_000,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "meta-llama/Llama-3.2-3B-Instruct",
name: "Llama 3.2 3B Instruct",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: false,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text],
cost: (0.050_9, 0.335, 0.0, 0.0),
context_window: 80_000,
max_tokens: 80_000,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "meta-llama/llama-3.3-70b-instruct",
name: "Llama 3.3 70B Instruct",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: false,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text],
cost: (1.04, 2.253, 0.0, 0.0),
context_window: 131_072,
max_tokens: 16_384,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "meta-llama/llama-4-maverick",
name: "Llama 4 Maverick",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: false,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (0.35, 1.15, 0.0, 0.0),
context_window: 1_048_576,
max_tokens: 115_200,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "minimax/minimax-m2.5",
name: "Minimax M2.5",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Off, None),
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, None),
(crate::types::ModelThinkingLevel::Medium, None),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text],
cost: (0.6, 2.4, 0.0, 0.0),
context_window: 204_800,
max_tokens: 128_000,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(false),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: Some(crate::types::ThinkingFormat::Openrouter),
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "minimax/minimax-m2.7",
name: "Minimax M2.7",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Off, None),
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, None),
(crate::types::ModelThinkingLevel::Medium, None),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text],
cost: (0.6, 2.4, 0.0, 0.0),
context_window: 204_800,
max_tokens: 131_072,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(false),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: Some(crate::types::ThinkingFormat::Openrouter),
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "minimax/minimax-m3",
name: "Minimax M3",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, None),
(crate::types::ModelThinkingLevel::Medium, None),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (0.3, 1.2, 0.0, 0.0),
context_window: 524_288,
max_tokens: 512_000,
featured: Some(true),
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(false),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: Some(crate::types::ThinkingFormat::Openrouter),
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "mistralai/mistral-large-2512",
name: "Mistral Large 2512",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: false,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (0.5, 1.5, 0.0, 0.0),
context_window: 262_144,
max_tokens: 209_715,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "mistralai/mistral-nemo",
name: "Mistral Nemo",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: false,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text],
cost: (0.165, 0.17, 0.0, 0.0),
context_window: 131_072,
max_tokens: 16_384,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "mistralai/mistral-small-2603",
name: "Mistral Small 2603",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, None),
(crate::types::ModelThinkingLevel::Medium, None),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (0.165, 0.66, 0.0, 0.0),
context_window: 262_144,
max_tokens: 209_715,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "mistralai/mixtral-8x22b-instruct",
name: "Mixtral 8X22B Instruct",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: false,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text],
cost: (2.2, 6.6, 0.0, 0.0),
context_window: 65_536,
max_tokens: 52_428,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "moonshotai/kimi-k2-0905",
name: "Kimi K2 0905",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: false,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text],
cost: (0.6, 2.5, 0.0, 0.0),
context_window: 98_304,
max_tokens: 98_304,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "moonshotai/kimi-k2.5",
name: "Kimi K2.5",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, None),
(crate::types::ModelThinkingLevel::Medium, None),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (0.6, 3.0, 0.0, 0.0),
context_window: 262_144,
max_tokens: 65_535,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(false),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: Some(crate::types::ThinkingFormat::Openrouter),
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "moonshotai/kimi-k2.6",
name: "Kimi K2.6",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, None),
(crate::types::ModelThinkingLevel::Medium, None),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (0.95, 4.0, 0.0, 0.0),
context_window: 262_144,
max_tokens: 262_144,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(false),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: Some(crate::types::ThinkingFormat::Openrouter),
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "moonshotai/kimi-k2.7-code",
name: "Kimi K2.7 Code",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Off, None),
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, None),
(crate::types::ModelThinkingLevel::Medium, None),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (1.9, 8.0, 0.0, 0.0),
context_window: 262_144,
max_tokens: 235_929,
featured: Some(true),
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(false),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: Some(crate::types::ThinkingFormat::Openrouter),
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "moonshotai/kimi-k3",
name: "Kimi K3",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, Some("low")),
(crate::types::ModelThinkingLevel::Medium, None),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, Some("max")),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (3.45, 17.25, 0.0, 0.0),
context_window: 1_048_576,
max_tokens: 1_048_576,
featured: Some(true),
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "nvidia/nemotron-3-nano-30b-a3b",
name: "Nemotron 3 Nano 30B A3B",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, None),
(crate::types::ModelThinkingLevel::Medium, None),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text],
cost: (0.06, 0.24, 0.0, 0.0),
context_window: 262_144,
max_tokens: 235_929,
featured: Some(true),
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(false),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: Some(crate::types::ThinkingFormat::Openrouter),
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "nvidia/nemotron-3-super-120b-a12b",
name: "Nemotron 3 Super 120B A12B",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, Some("low")),
(crate::types::ModelThinkingLevel::Medium, Some("medium")),
(crate::types::ModelThinkingLevel::High, None),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text],
cost: (0.3, 0.65, 0.0, 0.0),
context_window: 262_144,
max_tokens: 4_096,
featured: Some(true),
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "openai/gpt-4.1",
name: "GPT 4.1",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: false,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (2.0, 8.0, 0.0, 0.0),
context_window: 1_047_576,
max_tokens: 32_768,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "openai/gpt-4.1-mini",
name: "GPT 4.1 Mini",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: false,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (0.4, 1.6, 0.0, 0.0),
context_window: 1_047_576,
max_tokens: 32_768,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "openai/gpt-4.1-nano",
name: "GPT 4.1 Nano",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: false,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (0.1, 0.4, 0.0, 0.0),
context_window: 1_047_576,
max_tokens: 32_768,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "openai/gpt-4o",
name: "GPT 4O",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: false,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (2.5, 10.0, 0.0, 0.0),
context_window: 128_000,
max_tokens: 16_384,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "openai/gpt-4o-mini",
name: "GPT 4O Mini",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: false,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (0.15, 0.6, 0.0, 0.0),
context_window: 128_000,
max_tokens: 16_384,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "openai/gpt-5",
name: "GPT 5",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Off, None),
(crate::types::ModelThinkingLevel::Minimal, Some("minimal")),
(crate::types::ModelThinkingLevel::Low, Some("low")),
(crate::types::ModelThinkingLevel::Medium, Some("medium")),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (1.25, 10.0, 0.0, 0.0),
context_window: 400_000,
max_tokens: 128_000,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "openai/gpt-5-mini",
name: "GPT 5 Mini",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Off, None),
(crate::types::ModelThinkingLevel::Minimal, Some("minimal")),
(crate::types::ModelThinkingLevel::Low, Some("low")),
(crate::types::ModelThinkingLevel::Medium, Some("medium")),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (0.25, 2.0, 0.0, 0.0),
context_window: 400_000,
max_tokens: 128_000,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "openai/gpt-5-nano",
name: "GPT 5 Nano",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Off, None),
(crate::types::ModelThinkingLevel::Minimal, Some("minimal")),
(crate::types::ModelThinkingLevel::Low, Some("low")),
(crate::types::ModelThinkingLevel::Medium, Some("medium")),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (0.05, 0.4, 0.0, 0.0),
context_window: 400_000,
max_tokens: 128_000,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "openai/gpt-5.1",
name: "GPT 5.1",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, Some("low")),
(crate::types::ModelThinkingLevel::Medium, Some("medium")),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (1.25, 10.0, 0.0, 0.0),
context_window: 400_000,
max_tokens: 128_000,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "openai/gpt-5.2",
name: "GPT 5.2",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, Some("low")),
(crate::types::ModelThinkingLevel::Medium, Some("medium")),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, Some("xhigh")),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (1.75, 14.0, 0.0, 0.0),
context_window: 400_000,
max_tokens: 128_000,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "openai/gpt-5.2-pro",
name: "GPT 5.2 PRO",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Off, None),
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, None),
(crate::types::ModelThinkingLevel::Medium, Some("medium")),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, Some("xhigh")),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (21.0, 168.0, 0.0, 0.0),
context_window: 400_000,
max_tokens: 128_000,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "openai/gpt-5.3-codex",
name: "GPT 5.3 Codex",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, Some("low")),
(crate::types::ModelThinkingLevel::Medium, Some("medium")),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, Some("xhigh")),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (1.75, 14.0, 0.0, 0.0),
context_window: 400_000,
max_tokens: 128_000,
featured: Some(true),
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "openai/gpt-5.4",
name: "GPT 5.4",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, Some("low")),
(crate::types::ModelThinkingLevel::Medium, Some("medium")),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, Some("xhigh")),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (2.5, 15.0, 0.0, 0.0),
context_window: 1_050_000,
max_tokens: 128_000,
featured: Some(true),
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "openai/gpt-5.4-mini",
name: "GPT 5.4 Mini",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, Some("low")),
(crate::types::ModelThinkingLevel::Medium, Some("medium")),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, Some("xhigh")),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (0.75, 4.5, 0.0, 0.0),
context_window: 400_000,
max_tokens: 128_000,
featured: Some(true),
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "openai/gpt-5.4-nano",
name: "GPT 5.4 Nano",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, Some("low")),
(crate::types::ModelThinkingLevel::Medium, Some("medium")),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, Some("xhigh")),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (0.2, 1.25, 0.0, 0.0),
context_window: 400_000,
max_tokens: 128_000,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "openai/gpt-5.4-pro",
name: "GPT 5.4 PRO",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Off, None),
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, None),
(crate::types::ModelThinkingLevel::Medium, Some("medium")),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, Some("xhigh")),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (30.0, 180.0, 0.0, 0.0),
context_window: 1_050_000,
max_tokens: 128_000,
featured: Some(true),
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "openai/gpt-5.5",
name: "GPT 5.5",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, Some("low")),
(crate::types::ModelThinkingLevel::Medium, Some("medium")),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, Some("xhigh")),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (5.0, 30.0, 0.0, 0.0),
context_window: 1_050_000,
max_tokens: 128_000,
featured: Some(true),
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "openai/gpt-5.6",
name: "GPT 5.6",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Xhigh, Some("xhigh")),
(crate::types::ModelThinkingLevel::Max, Some("max")),
],
input: &[crate::types::ModelInput::Text],
cost: (5.0, 30.0, 0.0, 0.0),
context_window: 128_000,
max_tokens: 8_192,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "openai/gpt-5.6-luna",
name: "GPT 5.6 Luna",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, Some("low")),
(crate::types::ModelThinkingLevel::Medium, Some("medium")),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, Some("xhigh")),
(crate::types::ModelThinkingLevel::Max, Some("max")),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (0.2, 1.2, 0.0, 0.0),
context_window: 1_050_000,
max_tokens: 128_000,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "openai/gpt-5.6-luna-pro",
name: "GPT 5.6 Luna PRO",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, Some("low")),
(crate::types::ModelThinkingLevel::Medium, Some("medium")),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, Some("xhigh")),
(crate::types::ModelThinkingLevel::Max, Some("max")),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (1.0, 6.0, 0.0, 0.0),
context_window: 1_050_000,
max_tokens: 128_000,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "openai/gpt-5.6-sol",
name: "GPT 5.6 SOL",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, Some("low")),
(crate::types::ModelThinkingLevel::Medium, Some("medium")),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, Some("xhigh")),
(crate::types::ModelThinkingLevel::Max, Some("max")),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (5.0, 30.0, 0.0, 0.0),
context_window: 1_050_000,
max_tokens: 128_000,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "openai/gpt-5.6-sol-pro",
name: "GPT 5.6 SOL PRO",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, Some("low")),
(crate::types::ModelThinkingLevel::Medium, Some("medium")),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, Some("xhigh")),
(crate::types::ModelThinkingLevel::Max, Some("max")),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (5.0, 30.0, 0.0, 0.0),
context_window: 1_050_000,
max_tokens: 128_000,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "openai/gpt-5.6-terra",
name: "GPT 5.6 Terra",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, Some("low")),
(crate::types::ModelThinkingLevel::Medium, Some("medium")),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, Some("xhigh")),
(crate::types::ModelThinkingLevel::Max, Some("max")),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (2.5, 15.0, 0.0, 0.0),
context_window: 1_050_000,
max_tokens: 128_000,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "openai/gpt-5.6-terra-pro",
name: "GPT 5.6 Terra PRO",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, Some("low")),
(crate::types::ModelThinkingLevel::Medium, Some("medium")),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, Some("xhigh")),
(crate::types::ModelThinkingLevel::Max, Some("max")),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (2.5, 15.0, 0.0, 0.0),
context_window: 1_050_000,
max_tokens: 128_000,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "openai/gpt-6-astra",
name: "GPT 6 Astra",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Off, None),
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, Some("low")),
(crate::types::ModelThinkingLevel::Medium, Some("medium")),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, Some("xhigh")),
(crate::types::ModelThinkingLevel::Max, Some("max")),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (10.0, 50.0, 0.0, 0.0),
context_window: 1_050_000,
max_tokens: 128_000,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "openai/gpt-oss-120b",
name: "GPT OSS 120B",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Off, None),
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, Some("low")),
(crate::types::ModelThinkingLevel::Medium, Some("medium")),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text],
cost: (0.35, 0.75, 0.0, 0.0),
context_window: 131_072,
max_tokens: 117_964,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "openai/gpt-oss-20b",
name: "GPT OSS 20B",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Off, None),
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, Some("low")),
(crate::types::ModelThinkingLevel::Medium, Some("medium")),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text],
cost: (0.07, 0.3, 0.0, 0.0),
context_window: 131_072,
max_tokens: 131_072,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "poolside/laguna-s-2.1",
name: "Laguna S 2.1",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, None),
(crate::types::ModelThinkingLevel::Medium, None),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text],
cost: (0.09, 0.18, 0.0, 0.0),
context_window: 1_048_576,
max_tokens: 131_072,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(false),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: Some(crate::types::ThinkingFormat::Openrouter),
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "poolside/laguna-xs-2.1",
name: "Laguna XS 2.1",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, None),
(crate::types::ModelThinkingLevel::Medium, None),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text],
cost: (0.06, 0.12, 0.0, 0.0),
context_window: 262_144,
max_tokens: 32_768,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(false),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: Some(crate::types::ThinkingFormat::Openrouter),
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "qwen/qwen3-235b-a22b-2507",
name: "QWEN3 235B A22B 2507",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: false,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text],
cost: (0.25, 1.0, 0.0, 0.0),
context_window: 262_144,
max_tokens: 16_384,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "qwen/qwen3-30b-a3b-instruct-2507",
name: "QWEN3 30B A3B Instruct 2507",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: false,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text],
cost: (0.13, 0.52, 0.0, 0.0),
context_window: 262_144,
max_tokens: 32_000,
featured: Some(true),
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "qwen/qwen3-8b",
name: "QWEN3 8B",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, None),
(crate::types::ModelThinkingLevel::Medium, None),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text],
cost: (0.117, 0.455, 0.0, 0.0),
context_window: 131_072,
max_tokens: 8_192,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(false),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: Some(crate::types::ThinkingFormat::Openrouter),
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "qwen/qwen3-coder",
name: "QWEN3 Coder",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: false,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text],
cost: (0.975, 4.875, 0.0, 0.0),
context_window: 262_144,
max_tokens: 65_536,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "qwen/qwen3-coder-next",
name: "QWEN3 Coder Next",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: false,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text],
cost: (0.3, 1.5, 0.0, 0.0),
context_window: 262_144,
max_tokens: 235_929,
featured: Some(true),
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "qwen/qwen3-max",
name: "QWEN3 MAX",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: false,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text],
cost: (1.2, 6.0, 0.0, 0.0),
context_window: 262_144,
max_tokens: 65_536,
featured: Some(true),
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "qwen/qwen3-vl-235b-a22b-instruct",
name: "QWEN3 VL 235B A22B Instruct",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: false,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (0.3, 1.5, 0.0, 0.0),
context_window: 262_144,
max_tokens: 32_768,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "qwen/qwen3-vl-235b-a22b-thinking",
name: "QWEN3 VL 235B A22B Thinking",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Off, None),
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, None),
(crate::types::ModelThinkingLevel::Medium, None),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (0.98, 4.0, 0.0, 0.0),
context_window: 131_072,
max_tokens: 32_768,
featured: Some(true),
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(false),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: Some(crate::types::ThinkingFormat::Openrouter),
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "qwen/qwen3-vl-30b-a3b-instruct",
name: "QWEN3 VL 30B A3B Instruct",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: false,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (0.2, 0.7, 0.0, 0.0),
context_window: 262_144,
max_tokens: 16_384,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "qwen/qwen3-vl-8b-instruct",
name: "QWEN3 VL 8B Instruct",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: false,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (0.117, 0.455, 0.0, 0.0),
context_window: 262_144,
max_tokens: 32_768,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "qwen/qwen3.5-35b-a3b",
name: "QWEN3.5 35B A3B",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, None),
(crate::types::ModelThinkingLevel::Medium, None),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (0.25, 1.3, 0.0, 0.0),
context_window: 262_144,
max_tokens: 16_384,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(false),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: Some(crate::types::ThinkingFormat::Openrouter),
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "qwen/qwen3.5-397b-a17b",
name: "QWEN3.5 397B A17B",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, None),
(crate::types::ModelThinkingLevel::Medium, None),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (0.6, 3.6, 0.0, 0.0),
context_window: 262_144,
max_tokens: 65_536,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(false),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: Some(crate::types::ThinkingFormat::Openrouter),
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "qwen/qwen3.6-27b",
name: "QWEN3.6 27B",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, None),
(crate::types::ModelThinkingLevel::Medium, None),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (0.6, 3.6, 0.0, 0.0),
context_window: 262_144,
max_tokens: 65_536,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(false),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: Some(crate::types::ThinkingFormat::Openrouter),
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "qwen/qwen3.6-35b-a3b",
name: "QWEN3.6 35B A3B",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, None),
(crate::types::ModelThinkingLevel::Medium, None),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (0.25, 1.25, 0.0, 0.0),
context_window: 262_144,
max_tokens: 235_929,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(false),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: Some(crate::types::ThinkingFormat::Openrouter),
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "qwen/qwen3.7-flash",
name: "QWEN3.7 Flash",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, None),
(crate::types::ModelThinkingLevel::Medium, None),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (0.03, 0.13, 0.0, 0.0),
context_window: 1_000_000,
max_tokens: 65_536,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(false),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: Some(crate::types::ThinkingFormat::Openrouter),
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "qwen/qwen3.8-max",
name: "QWEN3.8 MAX",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Off, None),
(crate::types::ModelThinkingLevel::Minimal, Some("minimal")),
(crate::types::ModelThinkingLevel::Low, Some("low")),
(crate::types::ModelThinkingLevel::Medium, Some("medium")),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, Some("xhigh")),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (2.0, 6.0, 0.0, 0.0),
context_window: 1_000_000,
max_tokens: 131_072,
featured: Some(true),
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "x-ai/grok-4.20",
name: "Grok 4.20",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, None),
(crate::types::ModelThinkingLevel::Medium, None),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (2.0, 6.0, 0.0, 0.0),
context_window: 2_000_000,
max_tokens: 30_000,
featured: Some(true),
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(false),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: Some(crate::types::ThinkingFormat::Openrouter),
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "x-ai/grok-4.20-multi-agent",
name: "Grok 4.20 Multi Agent",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Off, None),
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, Some("low")),
(crate::types::ModelThinkingLevel::Medium, Some("medium")),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, Some("xhigh")),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (2.0, 6.0, 0.0, 0.0),
context_window: 2_000_000,
max_tokens: 30_000,
featured: Some(true),
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "x-ai/grok-4.6",
name: "Grok 4.6",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Off, None),
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, Some("low")),
(crate::types::ModelThinkingLevel::Medium, Some("medium")),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, Some("xhigh")),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (2.0, 6.0, 0.0, 0.0),
context_window: 500_000,
max_tokens: 450_000,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(true),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "xiaomi/mimo-v2.5",
name: "Mimo V2.5",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, None),
(crate::types::ModelThinkingLevel::Medium, None),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (0.168, 0.336, 0.0, 0.0),
context_window: 1_050_000,
max_tokens: 131_072,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(false),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: Some(crate::types::ThinkingFormat::Openrouter),
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "xiaomi/mimo-v2.5-pro",
name: "Mimo V2.5 PRO",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, None),
(crate::types::ModelThinkingLevel::Medium, None),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text],
cost: (0.480_24, 1.5, 0.0, 0.0),
context_window: 1_050_000,
max_tokens: 131_072,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(false),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: Some(crate::types::ThinkingFormat::Openrouter),
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "z-ai/glm-4.5",
name: "GLM 4.5",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, None),
(crate::types::ModelThinkingLevel::Medium, None),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text],
cost: (0.6, 2.2, 0.0, 0.0),
context_window: 131_072,
max_tokens: 98_304,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(false),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "z-ai/glm-4.5-air",
name: "GLM 4.5 AIR",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, None),
(crate::types::ModelThinkingLevel::Medium, None),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text],
cost: (0.2, 1.1, 0.0, 0.0),
context_window: 131_072,
max_tokens: 98_304,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(false),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "z-ai/glm-4.6",
name: "GLM 4.6",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, None),
(crate::types::ModelThinkingLevel::Medium, None),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text],
cost: (0.6, 2.2, 0.0, 0.0),
context_window: 204_800,
max_tokens: 131_072,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(false),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "z-ai/glm-4.7",
name: "GLM 4.7",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, None),
(crate::types::ModelThinkingLevel::Medium, None),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text],
cost: (0.6, 2.2, 0.0, 0.0),
context_window: 204_800,
max_tokens: 131_072,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(false),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "z-ai/glm-4.7-flash",
name: "GLM 4.7 Flash",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, None),
(crate::types::ModelThinkingLevel::Medium, None),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text],
cost: (0.07, 0.4, 0.0, 0.0),
context_window: 202_752,
max_tokens: 16_384,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(false),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "z-ai/glm-5",
name: "GLM 5",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, None),
(crate::types::ModelThinkingLevel::Medium, None),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text],
cost: (1.0, 3.2, 0.0, 0.0),
context_window: 204_800,
max_tokens: 131_072,
featured: Some(true),
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(false),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "z-ai/glm-5.1",
name: "GLM 5.1",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, None),
(crate::types::ModelThinkingLevel::Medium, None),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text],
cost: (1.4, 4.4, 0.0, 0.0),
context_window: 204_800,
max_tokens: 128_000,
featured: Some(true),
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(false),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "z-ai/glm-5.2",
name: "GLM 5.2",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, None),
(crate::types::ModelThinkingLevel::Medium, None),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, Some("xhigh")),
(crate::types::ModelThinkingLevel::Max, None),
],
input: &[crate::types::ModelInput::Text],
cost: (1.54, 4.84, 0.0, 0.0),
context_window: 1_048_576,
max_tokens: 131_072,
featured: Some(true),
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(false),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "z-ai/glm-5.3",
name: "GLM 5.3",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Off, None),
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, Some("low")),
(crate::types::ModelThinkingLevel::Medium, None),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, Some("max")),
],
input: &[crate::types::ModelInput::Text],
cost: (1.4, 4.4, 0.0, 0.0),
context_window: 1_310_720,
max_tokens: 262_144,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(false),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "prime-inference",
id: "z-ai/glm-5.3-flash",
name: "GLM 5.3 Flash",
api: "openai-completions",
base_url: "https://api.pinference.ai/api/v1",
reasoning: true,
thinking_levels: &[
(crate::types::ModelThinkingLevel::Off, None),
(crate::types::ModelThinkingLevel::Minimal, None),
(crate::types::ModelThinkingLevel::Low, Some("low")),
(crate::types::ModelThinkingLevel::Medium, None),
(crate::types::ModelThinkingLevel::High, Some("high")),
(crate::types::ModelThinkingLevel::Xhigh, None),
(crate::types::ModelThinkingLevel::Max, Some("max")),
],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (0.15, 0.5, 0.0, 0.0),
context_window: 1_310_720,
max_tokens: 131_072,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: Some(false),
supports_developer_role: Some(false),
supports_reasoning_effort: Some(false),
max_tokens_field: Some(crate::types::MaxTokensField::MaxTokens),
requires_reasoning_content_on_assistant_messages: None,
thinking_format: None,
zai_tool_stream: None,
send_session_affinity_headers: None,
supports_strict_mode: Some(false),
}),
},
RawModel {
provider: "vercel-ai-gateway",
id: "alibaba/qwen-3-14b",
name: "Qwen3-14B",
api: "anthropic-messages",
base_url: "https://ai-gateway.vercel.sh",
reasoning: true,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text],
cost: (0.12, 0.24, 0.0, 0.0),
context_window: 40_960,
max_tokens: 16_384,
featured: None,
headers: &[],
compat: None,
},
RawModel {
provider: "xai",
id: "grok-4.20-0309-non-reasoning",
name: "Grok 4.20 (Non-Reasoning)",
api: "openai-completions",
base_url: "https://api.x.ai/v1",
reasoning: false,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text, crate::types::ModelInput::Image],
cost: (1.25, 2.5, 0.2, 0.0),
context_window: 1_000_000,
max_tokens: 30_000,
featured: None,
headers: &[],
compat: None,
},
RawModel {
provider: "xiaomi",
id: "mimo-v2-flash",
name: "MiMo-V2-Flash",
api: "anthropic-messages",
base_url: "https://api.xiaomimimo.com/anthropic",
reasoning: true,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text],
cost: (0.14, 0.28, 0.002_8, 0.0),
context_window: 262_144,
max_tokens: 65_536,
featured: None,
headers: &[],
compat: None,
},
RawModel {
provider: "xiaomi-token-plan-ams",
id: "mimo-v2-flash",
name: "MiMo-V2-Flash",
api: "anthropic-messages",
base_url: "https://token-plan-ams.xiaomimimo.com/anthropic",
reasoning: true,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text],
cost: (0.14, 0.28, 0.002_8, 0.0),
context_window: 262_144,
max_tokens: 65_536,
featured: None,
headers: &[],
compat: None,
},
RawModel {
provider: "xiaomi-token-plan-cn",
id: "mimo-v2-flash",
name: "MiMo-V2-Flash",
api: "anthropic-messages",
base_url: "https://token-plan-cn.xiaomimimo.com/anthropic",
reasoning: true,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text],
cost: (0.14, 0.28, 0.002_8, 0.0),
context_window: 262_144,
max_tokens: 65_536,
featured: None,
headers: &[],
compat: None,
},
RawModel {
provider: "xiaomi-token-plan-sgp",
id: "mimo-v2-flash",
name: "MiMo-V2-Flash",
api: "anthropic-messages",
base_url: "https://token-plan-sgp.xiaomimimo.com/anthropic",
reasoning: true,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text],
cost: (0.14, 0.28, 0.002_8, 0.0),
context_window: 262_144,
max_tokens: 65_536,
featured: None,
headers: &[],
compat: None,
},
RawModel {
provider: "zai",
id: "glm-4.7",
name: "GLM-4.7",
api: "openai-completions",
base_url: "https://api.z.ai/api/coding/paas/v4",
reasoning: true,
thinking_levels: &[],
input: &[crate::types::ModelInput::Text],
cost: (0.0, 0.0, 0.0, 0.0),
context_window: 204_800,
max_tokens: 131_072,
featured: None,
headers: &[],
compat: Some(RawCompat {
supports_store: None,
supports_developer_role: Some(false),
supports_reasoning_effort: None,
max_tokens_field: None,
requires_reasoning_content_on_assistant_messages: None,
thinking_format: Some(crate::types::ThinkingFormat::Zai),
zai_tool_stream: Some(true),
send_session_affinity_headers: None,
supports_strict_mode: None,
}),
},];
/// Expand a raw compat into the wire compat object (the fallback
/// table only uses the OpenAI-completions shape).
fn compat_from_raw(raw: &RawCompat) -> ModelCompat {
ModelCompat::from_kind(CompatKind::OpenAiCompletions(Box::new(
OpenAiCompletionsCompat {
supports_store: raw.supports_store,
supports_developer_role: raw.supports_developer_role,
supports_reasoning_effort: raw.supports_reasoning_effort,
max_tokens_field: raw.max_tokens_field,
requires_reasoning_content_on_assistant_messages: raw
.requires_reasoning_content_on_assistant_messages,
thinking_format: raw.thinking_format,
zai_tool_stream: raw.zai_tool_stream,
send_session_affinity_headers: raw.send_session_affinity_headers,
supports_strict_mode: raw.supports_strict_mode,
..OpenAiCompletionsCompat::default()
},
)))
}
/// Expand one raw table entry into a full [`Model`].
fn model_from_raw(raw: &RawModel) -> Model {
let thinking_level_map = (!raw.thinking_levels.is_empty()).then(|| {
raw.thinking_levels
.iter()
.copied()
.map(|(level, value)| (level, value.map(String::from)))
.collect()
});
let headers = (!raw.headers.is_empty()).then(|| {
raw.headers
.iter()
.copied()
.map(|(k, v)| (k.to_string(), v.to_string()))
.collect()
});
Model {
id: raw.id.to_string(),
name: raw.name.to_string(),
api: raw.api.to_string(),
provider: raw.provider.to_string(),
base_url: raw.base_url.to_string(),
reasoning: raw.reasoning,
thinking_level_map,
input: raw.input.to_vec(),
cost: ModelCost {
input: JsNumber::from(raw.cost.0),
output: JsNumber::from(raw.cost.1),
cache_read: JsNumber::from(raw.cost.2),
cache_write: JsNumber::from(raw.cost.3),
},
context_window: raw.context_window,
max_tokens: raw.max_tokens,
max_tokens_explicit: false,
featured: raw.featured,
headers,
compat: raw.compat.as_ref().map(compat_from_raw),
}
}
fn catalog() -> &'static ProviderCatalog {
static CATALOG: LazyLock<ProviderCatalog> = LazyLock::new(|| {
let mut catalog: ProviderCatalog = HashMap::with_capacity(RAW_MODELS.len());
for raw in RAW_MODELS {
let model = model_from_raw(raw);
catalog
.entry(model.provider.clone())
.or_default()
.insert(model.id.clone(), model);
}
catalog
});
&CATALOG
}
/// Look up a model by provider and model id (TS `getModel`).
#[must_use]
pub fn get_model(provider: &str, model_id: &str) -> Option<&'static Model> {
catalog().get(provider)?.get(model_id)
}
/// All providers in the catalog (TS `getProviders`).
pub fn get_providers() -> Vec<&'static str> {
let mut providers: Vec<&'static str> = catalog().keys().map(String::as_str).collect();
providers.sort_unstable();
providers
}
/// All models for one provider, in catalog order (TS `getModels`).
#[must_use]
pub fn get_models(provider: &str) -> Vec<&'static Model> {
let Some(models) = catalog().get(provider) else {
return Vec::new();
};
// The table is provider/id-sorted, but keep the explicit id sort so the
// ordering never depends on the HashMap iteration order.
let mut models: Vec<&'static Model> = models.values().collect();
models.sort_by(|a, b| a.id.cmp(&b.id));
models
}
#[cfg(test)]
mod tests {
use super::*;
use crate::types::ModelExt;
#[test]
fn catalog_loads_at_the_ts_fallback_scale() {
let providers = get_providers();
assert_eq!(providers.len(), 32, "the TS fallback provider list");
let total: usize = providers
.iter()
.map(|provider| get_models(provider).len())
.sum();
assert_eq!(total, 152, "the TS fallback model list");
assert_eq!(
get_models("prime-inference").len(),
110,
"offline Prime Inference onboarding entries"
);
// The compiled transport table: one-plus template per tuple.
// The models are collected first: the tuples borrow from them, so
// the values must outlive the loop.
let models: Vec<Model> = RAW_MODELS.iter().map(model_from_raw).collect();
let mut tuples: Vec<(&str, &str, &str)> = Vec::new();
for model in &models {
let tuple = (
model.provider.as_str(),
model.api.as_str(),
model.base_url.as_str(),
);
if !tuples.contains(&tuple) {
tuples.push(tuple);
}
}
assert_eq!(
tuples.len(),
43,
"42 public transport tuples + Prime Inference"
);
}
#[test]
fn looks_up_known_models() {
let model = get_model("anthropic", "claude-fable-5")
.or_else(|| get_model("amazon-bedrock", "eu.anthropic.claude-fable-5"))
.expect("fable template present");
assert!(model.reasoning);
assert!(model.context_window > 0);
assert!(model.cost.input.as_f64() >= 0.0);
}
#[test]
fn unknown_lookups_are_none() {
assert!(get_model("nope", "nope").is_none());
assert!(get_models("nope").is_empty());
}
/// The compiled Prime Inference section is the conservative fallback (TS #2519): it declares no
/// reasoning controls — the live catalog's `supported_parameters` rebuild them at runtime — and
/// it never carries the zai thinking format (TS #2459: Prime Inference rejects
/// `enable_thinking` with a 400).
#[test]
fn prime_inference_fallback_declares_no_reasoning_controls() {
let glm_53 = get_model("prime-inference", "z-ai/glm-5.3").expect("compiled");
let Some(crate::types::CompatKind::OpenAiCompletions(compat)) = glm_53.compat_kind() else {
panic!("glm-5.3 carries a compat object");
};
assert_eq!(compat.supports_reasoning_effort, Some(false));
assert_eq!(compat.thinking_format, None);
assert_eq!(
glm_53
.thinking_level_map
.as_ref()
.and_then(|map| map.get(&crate::types::ModelThinkingLevel::High)),
Some(&Some("high".to_string()))
);
for model in get_models("prime-inference") {
let Some(crate::types::CompatKind::OpenAiCompletions(compat)) = model.compat_kind()
else {
panic!("prime-inference model {} carries a compat object", model.id);
};
assert_ne!(
compat.thinking_format,
Some(crate::types::ThinkingFormat::Zai),
"prime-inference model {} must not carry the zai thinking format",
model.id
);
}
// The direct z.ai provider keeps the toggle.
let glm = get_model("zai", "glm-4.7").expect("direct zai template is compiled");
assert!(matches!(
glm.compat_kind(),
Some(crate::types::CompatKind::OpenAiCompletions(compat))
if compat.thinking_format == Some(crate::types::ThinkingFormat::Zai)
));
}
}