Launches · Project

LiteLLM

LiteLLM provides a Python SDK and self-hosted gateway for routing requests to model providers through a common API.

+ Follow

Builder facts

  • Routing. The router supports retries, fallbacks and strategies such as least-busy or weighted routing. Source ↗
  • Access controls. The proxy can issue virtual keys tied to allowed models, budgets and rate limits. The documented key setup uses PostgreSQL. Source ↗
  • Upstream providers. Hosted model calls remain subject to the selected provider’s credentials and billing. Source ↗
  • Licensing. The repository’s core code is MIT-licensed and the enterprise directory has separate license terms. Source ↗

Sources checked

Publisher update ·

LiteLLM 1.110.0 ↗

Adds sequence numbers to each streaming event when the gateway bridges a provider’s response to the Responses API.

Change timeline

0 stories

No verified releases in this timeline yet.