FreeLLM-MCP
OfficialServer Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| FREEMCP_HTTP_PORT | No | Port for the HTTP transport when FREEMCP_TRANSPORT=http. | |
| FREEMCP_TRANSPORT | No | Set to 'http' to serve Streamable HTTP on http://127.0.0.1:8901/mcp instead of stdio. | |
| FREEMCP_CACHE_PATH | No | Move the cache location. Default is five minutes cache. | |
| FREEMCP_REMOTE_URL | No | Override the source URL for the public board. Default is https://freellmwatch.xyz/api/status.json | |
| FREEMCP_STATUS_PATH | No | Point at a local status JSON file instead of the remote 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 | {
"listChanged": false
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| list_free_modelsA | List all tracked free LLM APIs with live measured status, throughput (tokens/sec), uptime, and tool-calling support. |
| get_fastest_free_modelA | Get the currently fastest healthy free LLM API by measured tokens/second. |
| get_most_reliable_free_modelA | Get the free LLM API with the highest measured 7-day uptime among those currently answering. |
| check_model_statusA | Get detailed live metrics and error history for a specific free model ID. |
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 4 tools
Each tool has a clearly distinct purpose: listing all models, fetching the fastest, fetching the most reliable, and checking status of a specific model. No overlap in function, so an agent can easily select the right tool.
All tool names follow a consistent verb_noun pattern in snake_case: list_free_models, get_fastest_free_model, get_most_reliable_free_model, check_model_status. The pattern is predictable and uniform.
With exactly 4 tools, the server is well-scoped for a read-only model directory. Each tool serves a distinct query need without redundancy or bloat.
The surface covers listing, filtered retrieval (fastest/reliable), and per-model status. Missing an explicit search-by-name tool, but check_model_status covers that by ID. No lifecycle operations are needed for a read-only directory, so minor gap is acceptable.