Skip to main content
Glama

Hugging Face MCP Server

by evalstate

Server Configuration

Describes the environment variables required to run the server.

NameRequiredDescriptionDefault
HF_TOKENNoHugging Face token used if DEFAULT_HF_TOKEN is not set (when running with stdio transport)
TRANSPORTNoThe transport type to use (stdio, sse, streamableHttp, or streamableHttpJson)streamableHttpJson
HF_API_TIMEOUTNoTimeout for Hugging Face API requests in milliseconds12500
USER_CONFIG_APINoURL to use for User settings (defaults to Local front-end)
DEFAULT_HF_TOKENNoDefault Hugging Face token used if no Authorization header is sent. Only set this in Development/Test environments or for local STDIO Deployments.
MCP_PING_ENABLEDNoEnable ping keep-alive for sessionstrue
MCP_PING_INTERVALNoInterval between ping cycles (in milliseconds)30000
MCP_STRICT_COMPLIANCENoSet to True for GET 405 rejects in JSON Mode (default serves a welcome page)
MCP_CLIENT_CONNECTION_CHECKNoHow often to check for stale sessions (in milliseconds)90000
MCP_CLIENT_CONNECTION_TIMEOUTNoRemove sessions inactive for this duration (in milliseconds)300000
MCP_CLIENT_HEARTBEAT_INTERVALNoHow often to check SSE connection health (in milliseconds)30000

Schema

Prompts

Interactive templates invoked by user choice

NameDescription

No prompts

Resources

Contextual data attached and managed by the client

NameDescription

No resources

Tools

Functions exposed to the LLM to take actions

NameDescription

No tools

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/evalstate/hf-mcp-server'

If you have feedback or need assistance with the MCP directory API, please join our Discord server