$ loading_
Use LiteLLM proxy tools for completions, embeddings, images, and admin tasks.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "litellm-mcp" yet — see the docs or source repo.
Use litellm-mcp to call a chat model through the LiteLLM proxy with the prompt: "Summarize what MCP tools do in three bullet points," and return a concise English answer.
A short structured response generated by the proxied model.
Use litellm-mcp to generate an embedding for this text: "Customer feedback: they want faster search and more accurate answers," and return the vector result with model details.
A vector representation of the text plus basic information about the embedding model used.
Use litellm-mcp to call image generation and create an image of "a minimalist futuristic office desk illustration, blue and white palette, isometric view," then return the image URL or result.
An image generated from the prompt, with an accessible result link or metadata.
Manage LiteLLM proxy users, keys, and spend logs through MCP tools.
Route LLM completion requests to OpenAI-compatible providers through MCP tools.
Proxy multiple MCP servers while reducing token usage with on-demand tool loading.
Delegate low-risk tasks to a cheaper model with main-agent review.
Manage multiple MCP servers and load tool schemas on demand efficiently.
Connect LightRAG through MCP for unified retrieval and knowledge QA integration.