LiteLLM simplifies model access, spend tracking and fallbacks across 100+ LLMs. It provides a unified API to access models from OpenAI, Azure, Gemini, Bedrock, Anthropic, and more, all in the OpenAI format.
The platform enables developers to easily manage model access, implement rate limiting, track costs, and handle fallbacks across multiple providers. It supports on-prem deployment and integrates with observability tools like Langfuse, Arize Phoenix, Langsmith, and OpenTelemetry.
- OpenAI-compatible API for 100+ LLM providers
- Cost tracking and budgeting by team, user, or project
- Load balancing, rate limiting, and model fallbacks
- Enterprise-ready with JWT, SSO, audit logs, and air-gapped support