litellm
Python SDK, Proxy Server (AI Gateway) to call 100+ LLM APIs in OpenAI (or native) format, with cost tracking, guardrails, loadbalancing and logging. [Bedrock, Azure, OpenAI, VertexAI, Cohere, Anthropic, Sagemaker, HuggingFace, VLLM, NVIDIA NIM]
TL;DR · 30-second scan
litellm (Python) — Python SDK, Proxy Server (AI Gateway) to call 100+ LLM APIs in OpenAI (or native) format, with cost tracking, guardrails, loadbalancing and logging. [Bedrock, Azure, OpenAI, VertexAI, Cohere, Anthropic, Sagemaker, HuggingFace, VLLM, NVIDIA NIM]
You're calling multiple LLM providers and want a unified interface plus cost tracking.
AI & ML
Universal proxy/SDK for any LLM API — OpenAI, Anthropic, Gemini, Mistral, Cohere, Bedrock, local Ollama, 100+ providers. Same interface for all. Built-in spend tracking, rate limiting, caching. The right pick if you want to abstract LLM provider lock-in.
You're calling multiple LLM providers and want a unified interface plus cost tracking.
You're only using one provider — direct SDK is simpler.
Add this badge to your README to show your project is curated on StackPicks. Free, lightweight (180×28 SVG), and gives your visitors a one-click way to see honest take + alternatives.
[](https://stackpicks.dev/repo/berriai-litellm)
<a href="https://stackpicks.dev/repo/berriai-litellm"><img src="https://stackpicks.dev/api/badge/berriai-litellm" alt="Featured on StackPicks" width="180" height="28" /></a>
Are you the maintainer of BerriAI/litellm? Add the badge and we'll feature your project in the next weekly newsletter (~2,000 builders).