Hugging Face MCP Server
Lets assistants search and use Hugging Face models, datasets, Spaces, and papers from chat.
Publisher
Hugging Face
Product Details
The Hugging Face MCP Server lets assistants search, inspect, and use the Hugging Face Hub directly from chat. Developers and researchers can discover models, datasets, Spaces, documentation, and research papers, fetch detailed repo metadata, preview dataset structure, create or duplicate repositories, and invoke Gradio Spaces.
It also runs authenticated Hugging Face Jobs for long-running workflows. Teams use it to find the best models or datasets for a task, research ML papers, answer implementation questions from the documentation, generate or transform media with Spaces, and manage remote compute jobs for dataset analysis, evaluation, or batch inference.
Key Use Cases
Open-Source AI Model Discovery and Evaluation
Enables machine learning engineers to search models, benchmark papers, and datasets on Hugging Face, inspecting card metadata and sample splits to select base architectures for enterprise projects.
Serverless Compute Job and Space Invocation
Allows AI developers to trigger remote training or evaluation jobs on Hugging Face compute infrastructure and dynamically invoke multimodal Gradio Spaces directly within automated pipelines.
Explore detailed deployment path
Requires Gemini. Access integration prerequisites, specialized agent configuration guides, and implementation documentation.