LiteLLM
Universal LLM proxy — call 100+ LLMs (OpenAI, Anthropic, Bedrock, Vertex) with one API.
Overview
LiteLLM: Universal LLM Proxy
LiteLLM is Universal LLM proxy — 100+ providers via OpenAI-format API. Locking your codebase to one LLM vendor's SDK is a strategic mistake. LiteLLM gives you one OpenAI-format API across 100+ providers — switch models in one line.
Key Features
- Universal LLM proxy — 100+ providers via OpenAI-format API
- Cost tracking, fallbacks, retries, observability
- 14K+ GitHub stars; Series A backed
- Used by every major agent framework
- Open source + cloud platform
Ideal Use Case
Engineering teams running LLM workloads across multiple providers — for cost optimization, redundancy, model A/B testing, or vendor independence.
Why Use LiteLLM
Locking your codebase to one LLM vendor's SDK is a strategic mistake. LiteLLM gives you one OpenAI-format API across 100+ providers — switch models in one line.
FAQ
Q: vs OpenRouter? A: Both unify provider APIs. LiteLLM is OSS-first, can self-host; OpenRouter is API-first.
Q: Free? A: OSS core is free; cloud platform has tiers.
tl;dr
Universal OSS LLM proxy. 100+ providers. 14K stars. The default for multi-LLM products.
Related
Looking for more options? Browse the Developer Tools directory or read our best AI coding tools listicle. LiteLLM is also tracked on Crunchbase.
Why Use LiteLLM
User Reviews
Similar Tools
