Hugging Face Hub and Data Management
Unified workflow for managing models, datasets, and spaces on the Hugging Face Hub, and exploring dataset contents via the Dataset Viewer. Used by ML engineers, data scientists, and platform administrators for model discovery, dataset curation, and repository management.
What You Can Do
MCP Tools
list-models
List and search ML models on the Hugging Face Hub.
get-model
Get detailed information about a specific model.
get-model-revision
Get model information at a specific revision or branch.
list-datasets
List and search datasets on the Hub.
get-dataset
Get detailed information about a specific dataset.
get-dataset-parquet
Get parquet file URLs for a dataset.
list-spaces
List and search Spaces on the Hub.
get-space
Get detailed information about a specific Space.
create-repo
Create a new model, dataset, or Space repository on the Hub.
delete-repo
Delete a repository from the Hub.
update-repo-settings
Update settings for a repository.
whoami
Get information about the authenticated user.
list-model-tags
List all available model tags grouped by type.
list-dataset-tags
List all available dataset tags grouped by type.
check-dataset-validity
Check if a dataset is valid and available on the Hub.
get-dataset-splits
Get the list of splits for a dataset.
preview-dataset-rows
Get the first rows of a dataset split for preview.
get-dataset-rows
Get rows from a dataset split with pagination.
search-dataset
Search for rows matching a query in a dataset.
filter-dataset
Filter dataset rows using a condition.
get-dataset-size
Get size information for a dataset.
get-dataset-statistics
Get statistical information about a dataset split.
get-croissant-metadata
Get Croissant metadata for a dataset.