TL;DR
An open-source AI gateway framework that provides a unified OpenAI-compatible interface to call 100+ LLM providers.
Key features
Unified API: Call 100+ LLMs like OpenAI, Anthropic, Gemini, and Bedrock through a single interface.
Flexible Deployment: Integrate directly as a Python SDK or deploy as a centralized proxy server (AI Gateway) for teams/organizations.
Enterprise Features: Includes built-in cost tracking, load balancing, guardrails, and logging for enterprise environments.
When to use it
Use it when developing applications that call multiple LLM providers and want to abstract away differences in SDKs and request formats, or when you need to manage and track LLM usage and costs across a team.