LiteLLM
- About
- Open-source AI gateway to call 100+ model providers with one API. Open-source alternative to OpenRouter.
- Catalogue
- No. 091
- Category
- AI
- GitHub stars
- ★ 0
Overview
LiteLLM is an open-source AI gateway that gives you one interface, in the OpenAI format, for calling more than 100 model providers, including OpenAI, Anthropic, Gemini, Bedrock and Azure. It works as a Python SDK or as a self-hosted proxy server for a team or organisation.
What it does well
LiteLLM lets you switch model providers without rewriting your code. The proxy adds virtual keys, spend tracking, guardrails, load balancing and an admin dashboard, and supports chat, embeddings, images, audio, batches and more. It is self-hosted, with an optional enterprise tier.
Why it’s in the collection
LiteLLM solves a real headache for any team using several AI providers: one API, one place to track spend, and no lock-in to a single vendor. The core is MIT-licensed; only a separate enterprise folder has its own license.
On GitHub
- Stars
- 0
- Forks
- 0
- Commits
- 0
- Repository
- BerriAI/litellm
Languages
- Python 83.1%
- TypeScript 11.7%
- Rust 3.3%
- Other 1.9%
OpenRouter
Portkey
Ollama
