Hugging Face Hub Scraper: Models, Datasets & Spaces avatar

Hugging Face Hub Scraper: Models, Datasets & Spaces

Pricing

from $1.00 / 1,000 repositories

Go to Apify Store
Hugging Face Hub Scraper: Models, Datasets & Spaces

Hugging Face Hub Scraper: Models, Datasets & Spaces

Search Hugging Face models, datasets and Spaces: downloads, likes, trending score, task, library, license, languages, base models, arXiv papers and links. Filter by owner or task, sort by downloads or likes. Official public API.

Pricing

from $1.00 / 1,000 repositories

Rating

0.0

(0)

Developer

Pankaj Bhatia

Pankaj Bhatia

Maintained by Community

Actor stats

0

Bookmarked

2

Total users

1

Monthly active users

2 days ago

Last modified

Share

Search the Hugging Face Hub, the largest home for open AI models, datasets and demo apps. For each model, dataset or Space you get downloads, likes, trending score, license, languages, tags, linked arXiv papers, creation and update dates and a link. Models add the task (e.g. text-generation), library, base models and training datasets. Datasets add task categories, size, modalities and formats. Spaces add the SDK, title, description and app URL. Uses the official public Hub API. No login, no token.

Who is it for?

  • AI engineers and researchers comparing models for a task by downloads, likes and license.
  • VCs, analysts and journalists tracking which open models and organizations are trending.
  • Developer-tool and cloud companies finding popular models to support or integrate.
  • Dataset curators discovering datasets by topic, size and format.

How to use it

  1. Pick repository types (models, datasets, Spaces).
  2. Type search terms like llama or whisper, or leave empty for the top repositories overall.
  3. Optionally filter by owner (e.g. google) or task, choose sort by, and set max results.
  4. Add repository IDs to look up specific repos. Then click Start.

Input example

{"repoTypes": ["models"], "searchQueries": ["qwen"], "task": "text-generation", "sortBy": "downloads", "maxResultsPerSearch": 100}

Output example

{"type": "model", "id": "Qwen/Qwen3-0.6B", "owner": "Qwen", "downloads": 23833726, "likes": 1850, "license": "apache-2.0", "task": "text-generation", "library": "transformers", "baseModels": ["Qwen/Qwen3-0.6B-Base"], "arxivPapers": ["2505.09388"], "url": "https://huggingface.co/Qwen/Qwen3-0.6B"}

How much does it cost?

Pay-per-event: a small fee per repository returned, plus a tiny per-run start fee, with no subscription and no separate platform usage costs. For example, 1,000 repositories cost about $1. Searches with no results are not charged. Set Maximum cost per run to cap your spending.

Use it through the API, integrations or AI agents

Run it from the Apify API, on a schedule, or connect it to Make, Zapier, n8n or Google Sheets. AI agents can find and call it through the Apify MCP server. Output is clean JSON with the same fields on every run.

FAQ

Is it legal? Yes. The tool reads public metadata from Hugging Face's official public API, within its anonymous rate limit. It never downloads model weights, gated files or private repositories.

What do downloads mean? Hugging Face reports downloads over roughly the last 30 days.

Can I track trends? Yes. Schedule a daily run sorted by Trending and compare the results.

More tools from this developer