LiteLLM
06 model-router/Community / OSS/ACTIVE
LiteLLM is a self-hostable proxy that normalizes hundreds of LLM APIs behind an OpenAI-compatible endpoint.
What it does
- 01OpenAI-compatible proxy
- 02Self-host
- 03Budgets and routing
- 04Open source
Who it's for
Teams that want a gateway they operate themselves; used under OpenHands and many agent stacks.
Pricing
Free OSS self-host; Enterprise quote-based. List prices as checked against the vendor's pricing page on 2026-09-25; plans change often, so confirm before deciding.
caveat · before you adopt
Enterprise licence is quote-based and priced by gateway scale, not per token; SSO is free up to 5 users.
Other model routers (1)
More from Community / OSS

Aider
Community / OSS · cli-agent
Cline
Community / OSS · ide-extension
OpenCode
Community / OSS · cli-agent

OpenHands
Community / OSS · cli-agent
Model Context Protocol
Community / OSS · bridge
Continue
Community / OSS · ide-extension
Instinct
Community / OSS · bridge

Agent Network Protocol
Community / OSS · agent-communication
Head-to-head (1)
Frequently asked
What is LiteLLM?
LiteLLM is a self-hostable proxy that normalizes hundreds of LLM APIs behind an OpenAI-compatible endpoint.
Who is LiteLLM for?
Teams that want a gateway they operate themselves; used under OpenHands and many agent stacks.
How much does LiteLLM cost?
Free OSS self-host; Enterprise quote-based. List prices as checked against the vendor's pricing page on 2026-09-25; plans change often, so confirm before deciding.
Does LiteLLM run locally or in the cloud?
LiteLLM runs locally on your machine and in the vendor's cloud. Its interfaces are an API or protocol.
What are alternatives to LiteLLM?
Other model router entries in the catalog: OpenRouter. Different categories solve the same problem differently — see the Model routers & gateways category page for tradeoffs.
What should I know before adopting LiteLLM?
Enterprise licence is quote-based and priced by gateway scale, not per token; SSO is free up to 5 users.