LiteLLM MCP Server Bridge
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| LITELLM_API_KEY | Yes | Your LiteLLM API key for authentication | |
| LITELLM_BASE_URL | Yes | Your LiteLLM instance URL |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| chat_completionC | Generate chat completions using LiteLLM. Supports all models available in your LiteLLM instance. |
| completionB | Generate text completions (legacy endpoint) using LiteLLM. |
| list_modelsA | List all available models from your LiteLLM instance |
| health_checkA | Check the health status of your LiteLLM instance |
| create_embeddingB | Generate embeddings using LiteLLM |
| model_infoA | Get detailed information about a specific model |
| create_imageB | Generate images using LiteLLM (/images/generations) |
| create_speechA | Generate speech audio from text using LiteLLM (/audio/speech) |
| rerankB | Rerank documents based on a query using LiteLLM (/rerank) |
| key_generateB | Generate a new LiteLLM Proxy API key (/key/generate) |
| key_infoA | Get details and spend info for a LiteLLM Proxy API key (/key/info) |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 11 tools
Most tools target distinct resources (keys, models, health, embeddings, images, speech, rerank), but chat_completion and completion could be confused as both generate text. Descriptions help clarify that one is for chat and the other is a legacy endpoint.
Naming conventions are mixed: some follow verb_noun (list_models, create_embedding), others noun_verb (key_generate, model_info), and some are single nouns or verbs (completion, health_check, rerank). This inconsistency makes the tool surface harder to scan.
With 11 tools, the set is well-scoped for a LiteLLM proxy bridge, covering key management, model introspection, and a variety of generation endpoints without feeling bloated or overly sparse.
Core operations are covered: key generation/info, model listing/info, chat and text completions, embeddings, image, speech, and rerank. Minor gaps exist, such as no key deletion/update or audio transcription, but most typical workflows are supported.