LiteLLM Unified LLM Gateway and Proxy Server

LiteLLM is an open-source Python SDK and proxy server that provides a unified OpenAI-compatible interface to call 100+ LLM APIs including OpenAI, Anthropic, Azure, Bedrock, and more. It includes cost tracking, guardrails, load balancing, and virtual key management for production deployments.

Installation

Use the upstream install or setup path that matches your environment:

• Stable Release: Use docker images with the -stable tag. These have undergone 12 hour load tests, before being published. [More information about the release cycle here](https://docs.litellm.ai/docs/proxy/release_c...
• Run npm run dev to start the dashboard

Requirements and caveats from upstream:

• You can use LiteLLM through either the Proxy Server or Python SDK. Both give you a unified interface to access multiple LLMs (100+ LLMs). Choose the option that best fits your needs:
• <th style={{width: '43%'}}><strong><a href="https://docs.litellm.ai/docs/">LiteLLM Python SDK</a></strong></th>
• <td style={{width: '43%'}}>Use LiteLLM directly in your Python code</td>

Basic usage or getting-started notes:

• Run in Developer Mode

• uv run prisma generate

• Source: https://github.com/BerriAI/litellm

• Extracted from upstream docs: https://raw.githubusercontent.com/BerriAI/litellm/HEAD/README.md

Source

• Agent Skill Exchange