LiteLLM AI Gateway
Self-hosted gateway that gives applications an OpenAI-format interface to multiple language-model providers, with virtual keys and spend tracking.
About this listing
LiteLLM AI Gateway centralizes access to language-model providers behind an OpenAI-format interface. Teams can deploy its proxy as a shared service rather than integrate each provider separately. The project also provides a Python SDK for direct application integration.
The gateway includes virtual keys, spend tracking and load balancing, making it useful for teams coordinating model access across applications. This listing covers the self-hosted open-source gateway, not a hosted service or enterprise subscription.
Free refers to the MIT-licensed code outside the enterprise directory. Model-provider usage, hosting and infrastructure can cost money; enterprise-directory code has separate licensing.