LiteLLM

An AI gateway and Python SDK that calls more than 100 model providers through one OpenAI-style API, with cost tracking, budgets, load balancing and logging.

62.5k stars, 12.9k forks Python On GitHub since July 2023

Visit website View on GitHub How to run it

What is LiteLLM?

LiteLLM gives every model provider, including OpenAI, Anthropic, Gemini, Bedrock, Azure and self-hosted servers like vLLM, the same request format. Developers can use it as a library, while platform teams deploy the proxy server as a central gateway that hands out virtual keys, tracks spend per team, enforces budgets and rate limits, retries and balances load across deployments, and sends logs to observability tools. Its core is now written in Rust, with the Python SDK on top.

Good fit if

  • Several teams use different model providers and you need one place for keys, budgets and logs.
  • You want to switch providers or add fallbacks without rewriting application code.
  • You run your own models and want them behind the same API as hosted ones.

Look elsewhere if

Before you adopt it

Most of LiteLLM is MIT-licensed, but the enterprise directory has its own commercial license, and features such as SSO and advanced controls need a paid key. BerriAI asks contributors to sign a CLA. The gateway uses PostgreSQL to store keys and spend.

Drafted with AI from the project's README and repository data, then checked against that data on 30 October 2026. Report a correction

LiteLLM health: is it still maintained?

83 of 100, actively developed

  • Activity
    13,517 commits in the last 90 days, the latest on 1 November 2026.
    30 / 30
  • Contributors
    25 people wrote the latest 100 commits; the most active wrote 11% of them.
    19 / 20
  • Releases
    12 releases in the last 12 months; the latest, v1.106.6, on 28 October 2026.
    18 / 20
  • Replies to issues
    Too few recent issues from outside the team to measure, so this part gets a neutral score.
    10 / 20
  • Security practices
    OpenSSF Scorecard 5.7 of 10.
    6 / 10

Read from GitHub on 1 November 2026. How the score works

LiteLLM license: is it open source?

LiteLLM uses MIT, an open-source license approved by the OSI. Some features sit under a separate paid license inside the same repository. Contributors sign a contributor license agreement, which lets the owner relicense their work later.

  1. July 2023First published on GitHub.
  2. TodayMIT, with paid parts: license file assigns a different licence to part of the code; directory enterprise/; separate license enterprise/LICENSE.md. A contributor license agreement is required.

How to run LiteLLM yourself

Docker
Dockerfile or compose file in the repository
Compose file
docker-compose.yml: 3 services (postgres)
Setup effort
2 of 5, judged from the compose file
Kubernetes
Helm chart in the repository
Desktop and mobile
No app builds in GitHub releases
Interface languages
English only

Install instructions: docs.litellm.ai or the README on GitHub.

Alternatives to LiteLLM

  • Hermes AgentThe agent that grows with you 76 health · MIT · 259k stars
  • LibreChatEnhanced ChatGPT Clone: Features Agents, MCP, Skills, DeepSeek, Anthropic, AWS, OpenAI, Responses API,… 74 health · MIT · 46.6k stars
  • Open WebUIUser-friendly AI Interface (Supports Ollama, OpenAI API, ...) 76 health · Open WebUI License · 157k stars
  • vLLMA high-throughput and memory-efficient inference and serving engine for LLMs 84 health · Apache-2.0 · 96k stars
  • FastGPTFastGPT is a knowledge-based platform built on the LLMs, offers a comprehensive suite of out-of-the-box… 82 health · FastGPT Open Source License · 30k stars

All 36 projects in AI chat & local LLMs

Questions about LiteLLM

Is LiteLLM open source?

Yes. LiteLLM is published under MIT, an OSI-approved open-source license. Some features are kept under a separate paid license in the same repository.

Is LiteLLM still maintained?

Its health score is 83 of 100 (actively developed). It had 13,517 commits in the last 90 days, the latest on 1 November 2026, and its latest release came out on 28 October 2026.

How do I run LiteLLM?

You can run it with Docker, on Kubernetes with the Helm chart in the repository. The README on GitHub has the exact steps.

What are the alternatives to LiteLLM?

Close alternatives in the AI chat & local LLMs category include Hermes Agent, LibreChat, Open WebUI.