The fastest, litest AI Gateway. Rust core with Python SDK. Call 100+ LLM APIs in OpenAI (or native) format with cost tracking, guardrails, load balancing, and logging [Bedrock, Azure, OpenAI, Anthropic, OpenAI, VertexAI, vLLM, Nvidia NIM]
Python ai-gatewayanthropicazure-openaibedrockgatewaylangchainlitellmllmllm-gatewayllmopsmcp-gatewayopenaiopenai-proxyrustrust-aivertex-ai
from collections.abc import AsyncIterator, Iterator
from pydantic import BaseModel
from litellm.responses.utils import ResponsesAPIRequestUtils
from litellm.types.llms.openai import ResponsesAPIResponse
def response_id(response: object) -> object:
if isinstance(response, ResponsesAPIResponse):
return ResponsesAPIRequestUtils.decode_responses_api_response_id(response.id).get("response_id") or response.id
return response.get("id") if isinstance(response, dict) else None
def with_id(response: object, new_id: str) -> object:
if isinstance(response, BaseModel):
return response.model_copy(update={"id": new_id})
assert isinstance(response, dict)
return {**response, "id": new_id}
async def consume(stream: object) -> tuple[object, ...]:
if isinstance(stream, AsyncIterator):
return tuple([chunk async for chunk in stream])
assert isinstance(stream, Iterator)
return tuple(stream)