Community MCP server for Hugging Face, the leading ML model hub. Agents can search models and datasets, run inference on hosted models, explore model cards and documentation, manage Spaces, and access paper references. Essential for ML engineering workflows — model discovery, benchmarking, and building inference pipelines.
npx -y @huggingface/mcp-server
{
"mcpServers": {
"huggingface": {
"env": {
"HF_TOKEN": "your-hf-token"
},
"args": [
"-y",
"@huggingface/mcp-server"
],
"command": "npx"
}
}
}Get HF model details
{
"type": "object",
"required": [
"model_id"
],
"properties": {
"model_id": {
"type": "string"
}
}
}// Input
{
"model_id": "bert-base-uncased"
}// Output
{
"id": "bert-base-uncased",
"downloads": 50000000,
"pipeline_tag": "fill-mask"
}Search HuggingFace models
{
"type": "object",
"required": [
"query"
],
"properties": {
"task": {
"type": "string"
},
"query": {
"type": "string"
}
}
}// Input
{
"task": "text-classification",
"query": "sentiment analysis"
}// Output
{
"models": [
{
"id": "bert-base",
"downloads": 1000000
}
]
}Community MCP server for Hugging Face, the leading ML model hub. Agents can search models and datasets, run inference on hosted models, explore model cards and documentation, manage Spaces, and access paper references. Essential for ML engineering workflows — model discovery, benchmarking, and building inference pipelines.
Hugging Face provides 2 tools including get_model_info, search_models.
Hugging Face offers a free tier with 200 calls per month. For higher usage, paid plans start at €0/month.
You can install Hugging Face using the following command: npx -y @huggingface/mcp-server. After installation, add the provided config snippet to your Claude Desktop or Cursor configuration.
Hugging Face is listed under the AI & Machine Learning category in the AgentForge MCP registry.
Hugging Face has a current uptime of 99.85% with an average response time of 450ms.
To connect Hugging Face, click the "Connect Agent" button on this page to get the configuration snippet. Add it to your MCP client (Claude Desktop, Cursor, or any MCP-compatible tool). Your AI agent will then have access to all of Hugging Face's tools via the Model Context Protocol.