LiteLLM
Self-hosted gateway and SDK to call 100+ LLM APIs in OpenAI format
What it is
A Python SDK and a self-hosted proxy server that expose many LLM providers behind one OpenAI-style interface. The proxy acts as a central gateway for a team, so applications talk to one endpoint while providers are swapped behind it. For platform engineers standardising LLM access.
Good for
- Put one API endpoint in front of many providers
- Swap model vendors without changing app code
- Run a shared LLM gateway for a team
Install or run
uv tool install 'litellm[proxy]'Command as given in the project README. Read the repository before you run anything on your machine.
What you need
Python; API keys for the providers you route to
Sources
- https://github.com/BerriAI/litellm (project README, read 10 Nov 2026)
- https://api.github.com/repos/BerriAI/litellm (GitHub API, read 13 Nov 2026)
The summary is ours, written from the README. Spotted an error? See how we work.