MCP.so
Sign In

🤗 Hugging Face MCP Server 🤗

@shreyaskarnik

Repository

About 🤗 Hugging Face MCP Server 🤗

No overview available yet

Config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "huggingface-mcp-server": {
      "command": "npx",
      "args": [
        "-y",
        "@smithery/cli",
        "install",
        "@shreyaskarnik/huggingface-mcp-server",
        "--client",
        "claude"
      ]
    }
  }
}

Tools

10

Search models with filters for query, author, tags, and limit

Get detailed information about a specific model

Search datasets with filters

Get detailed information about a specific dataset

Search Spaces with filters including SDK type

Get detailed information about a specific Space

Get information about a paper and its implementations

Get the list of curated daily papers

Search collections with various filters

Get detailed information about a specific collection

Overview

What is Hugging Face MCP Server?

A Model Context Protocol (MCP) server that provides read-only access to the Hugging Face Hub APIs. It allows LLMs like Claude to interact with Hugging Face's models, datasets, spaces, papers, and collections.

How to use Hugging Face MCP Server?

Install via Smithery or configure the server in Claude Desktop’s claude_desktop_config.json using uv run huggingface_mcp_server.py. Set the optional HF_TOKEN environment variable for higher rate limits and private repository access. After installation, use the provided prompts and tools to query the Hub.

Key features of Hugging Face MCP Server

  • Custom hf:// URI scheme for models, datasets, and spaces
  • Prompts: compare models and summarize research papers
  • Tools to search and get info on models, datasets, spaces, papers, and collections
  • Optional Hugging Face authentication via HF_TOKEN
  • No required configuration; works out-of-the-box

Use cases of Hugging Face MCP Server

  • Search for models or datasets by query, author, or tags
  • Compare multiple Hugging Face models side by side
  • Summarize a research paper using its arXiv ID
  • Retrieve the list of curated daily papers
  • Explore collections and their contents

FAQ from Hugging Face MCP Server

What is the HF_TOKEN environment variable used for?

It is optional. Setting it provides higher API rate limits and access to private repositories (if authorized). It also improves reliability for high-volume requests.

How do I debug the server?

Use the MCP Inspector by running npx @modelcontextprotocol/inspector uv --directory /path/to/huggingface-mcp-server run huggingface_mcp_server.py. Server logs are stored in ~/Library/Logs/Claude/mcp-server-huggingface.log (macOS) or %APPDATA%\Claude\logs\mcp-server-huggingface.log (Windows).

How do I install for Claude Desktop?

You can install automatically via Smithery with npx -y @smithery/cli install @shreyaskarnik/huggingface-mcp-server --client claude, or manually add the configuration to claude_desktop_config.json under mcpServers.

What resources does the server expose?

It exposes models (hf://model/{model_id}), datasets (hf://dataset/{dataset_id}), spaces (hf://space/{space_id}), and daily papers and collections via tools. All resources have descriptive names and JSON content type.

Is the server read-only?

Yes. The server provides read-only access to the Hugging Face Hub APIs. It cannot modify or upload data.

Frequently asked questions

What is the `HF_TOKEN` environment variable used for?

It is optional. Setting it provides higher API rate limits and access to private repositories (if authorized). It also improves reliability for high-volume requests.

How do I debug the server?

Use the MCP Inspector by running `npx @modelcontextprotocol/inspector uv --directory /path/to/huggingface-mcp-server run huggingface_mcp_server.py`. Server logs are stored in `~/Library/Logs/Claude/mcp-server-huggingface.log` (macOS) or `%APPDATA%\Claude\logs\mcp-server-huggingface.log` (Windows).

How do I install for Claude Desktop?

You can install automatically via Smithery with `npx -y @smithery/cli install @shreyaskarnik/huggingface-mcp-server --client claude`, or manually add the configuration to `claude_desktop_config.json` under `mcpServers`.

What resources does the server expose?

It exposes models (`hf://model/{model_id}`), datasets (`hf://dataset/{dataset_id}`), spaces (`hf://space/{space_id}`), and daily papers and collections via tools. All resources have descriptive names and JSON content type.

Is the server read-only?

Yes. The server provides read-only access to the Hugging Face Hub APIs. It cannot modify or upload data.

Comments

Basic information

Category

AI & Agents

License

MIT

Transports

stdio

Author

shreyaskarnik

More AI & Agents MCP servers

mcp-hfspace MCP Server 🤗

evalstate

MCP Server to Use HuggingFace spaces, easy configuration and Claude Desktop mode.

387

Model Context Protocol for Unreal Engine

chongdashu

Enable AI assistant clients like Cursor, Windsurf and Claude Desktop to control Unreal Engine through natural language using the Model Context Protocol (MCP).

2.0K

bsv.cx

Andy Rowe

Timestamp and verify evidence on-chain. Let your agent prove what it saw and when. bsv.cx gives an AI agent verifiable memory. Timestamp a hash of any output, archive a web page as it looked, or fetch a URL with a recei

21st.dev Magic AI Agent

21st-dev

It's like v0 but in your Cursor/WindSurf/Cline. 21st dev Magic MCP server for working with your frontend like Magic

5.3K

Unreal Engine Generative AI Support Plugin

prajwalshettydev

Unreal Engine plugin for LLM/GenAI models & MCP UE5 server. OpenAI GPT-5, Deepseek R1, Claude Opus/Sonnet, Gemini 3, Grok 4, Alibaba Qwen, Kimi, ElevenLabs TTS, Inworld, OpenRouter, Groq, GLM, Ollama, Local, Meshy, Tripo, Hunyuan3D, Rodin, fal, Dashscope, Seedream. NPC AI, agenti

622