Search across 710 pages

Try a tool name, category, or "lifetime deal"

↑↓ navigate ↵ open Esc close
Live directory · Updated weekly · No pay-to-rank

Hugging Face Datasets: The Complete Directory & Live Ranking

A live directory of Hugging Face Hub datasets, ranked by real adoption, growth, and maintenance - not opinion.

Alston Antony Curated & maintained by Alston Antony · 15+ yrs SEO, 500+ tools tested
150Datasets ranked
174.6MCombined downloads
7Categories
Live data from the Hugging Face Hub API
Ranked by real adoption & growth
No dataset pays to be listed or ranked
See the full ranking methodology →
Updated August 4, 2026 Hugging Face Hub API 150 datasets ranked Refreshed weekly Independent & ad-free
What & how

What Is a Hugging Face Dataset, and How Do You Use One?

What it is

A Hugging Face dataset is a published collection of data (text, images, audio, or tabular) hosted on the Hugging Face Hub, along with a dataset card documenting its schema, splits, and license. Datasets are used to train, fine-tune, and evaluate AI models.

How to use a Hugging Face dataset

Every dataset on this list can be loaded directly with the datasets library (load_dataset("org/dataset-id")), which handles download, caching, and format conversion automatically. Click a dataset's name to open its official Hugging Face page, where the dataset card documents the exact schema, splits, and any license terms to accept.

Browse the full directory below ↓

As of August 4, 2026, this is a live, free directory ranking the best and most-used datasets on the Hugging Face Hub by real data. zplatform.ai tracks datasets that clear a minimum adoption floor (150 ranked today) using the free, public Hugging Face Hub API. Ranked by downloads, likes, and how recently each dataset was updated - not opinion - and the list refreshes automatically every week. How we choose and rank →

Also explore the ranked Hugging Face models directory and the best MCP servers for connecting data to AI apps.

The best Hugging Face datasets as of August 4, 2026, ranked by our composite score:

  1. assets - 764.4k downloads, quality score 90/100.
  2. HPLT2.0_cleaned - 1.6M downloads, quality score 90/100.
  3. samples - 577.0k downloads, quality score 90/100.
  4. seedance-2-prompts-datasets - 337.0k downloads, quality score 89/100.
  5. PhysicalAI-Robotics-GR00T-X-Embodiment-Sim - 12.6M downloads, quality score 89/100.

Every Ranked Dataset

Every Hugging Face dataset that clears our adoption floor, scored on a transparent composite of adoption, growth, maintenance, and trust. Select up to 4 to compare side by side.

Compare # Dataset Category Downloads Likes License Updated Quality
Editor's picks

Best Hugging Face Datasets by Use Case

Best Speech & Audio Datasets

WaxalNLP and zoengjyutgaai lead speech and audio datasets, used to train transcription and audio-classification models.

  1. WaxalNLP 88/100 quality
  2. zoengjyutgaai 85/100 quality
  3. fleurs 85/100 quality

Best Robotics & Simulation Datasets

PhysicalAI-Robotics-GR00T-X-Embodiment-Sim and L2D lead robotics and simulation datasets, a fast-growing category behind today's agentic and embodied-AI benchmarks.

  1. PhysicalAI-Robotics-GR00T-X-Embodiment-Sim 89/100 quality
  2. L2D 87/100 quality
  3. PhysicalAI-Robotics-Open-H-Embodiment 86/100 quality

Best Text & Language Datasets

HPLT2.0_cleaned and Ultra-FineWeb lead text and language datasets, covering generation, classification, translation, and embedding training data.

  1. HPLT2.0_cleaned 90/100 quality
  2. Ultra-FineWeb 88/100 quality
  3. stack-v3-train 87/100 quality

Best Vision & Video Datasets

seedance-2-prompts-datasets and OpenFake top vision and video datasets, spanning image generation, classification, and multi-frame understanding.

  1. seedance-2-prompts-datasets 89/100 quality
  2. OpenFake 88/100 quality
  3. opencs2_dataset 85/100 quality

Best Tabular & Time-Series Datasets

OpenAlex and USDT-M_Perpetual_Futures top tabular and time-series datasets.

  1. OpenAlex 85/100 quality
  2. USDT-M_Perpetual_Futures 84/100 quality
  3. MacroLens 82/100 quality

Best Question Answering & Reasoning Datasets

prompts.chat and dartlab-data top question-answering and reasoning benchmarks.

  1. prompts.chat 88/100 quality
  2. dartlab-data 83/100 quality
  3. MMMU 83/100 quality
Browse

Browse by Category

Text & Language Data

25 datasets · 76.3M downloads

Filter the directory to Text & Language Data ↓

Vision & Video

29 datasets · 9.4M downloads

Filter the directory to Vision & Video ↓

Robotics & Simulation

10 datasets · 14.7M downloads

Filter the directory to Robotics & Simulation ↓

Question Answering & Reasoning

6 datasets · 5.7M downloads

Filter the directory to Question Answering & Reasoning ↓

Speech & Audio

3 datasets · 2.1M downloads

Filter the directory to Speech & Audio ↓

Tabular & Time-Series

4 datasets · 436.9k downloads

Filter the directory to Tabular & Time-Series ↓

Other

73 datasets · 65.9M downloads

Filter the directory to Other ↓

State of the Datasets Ecosystem

Most Popular

  1. KakologArchives 29.5M
  2. gsm8k 14.1M
  3. PhysicalAI-Robotics-GR00T-X-Embodiment-Sim 12.6M
  4. SWE-rebench 12.4M
  5. fineweb-tokenized 8.8M
  6. ubuntu_osworld_file_cache 7.2M
  7. hd_tmp 6.3M
  8. openwebtext 5.5M
  9. flores 4.3M
  10. LLaVA-OneVision-1.5-Mid-Training-85M 3.5M

Fastest Growing (30-day downloads trend)

  1. video-vec2wav2-tokenizer +321.3%
  2. RoboDojo +190.1%
  3. 2026-challenge-demos +176.2%
  4. SWE-rebench +156.3%
  5. 4D-Lung +147.3%
  6. opencs2_dataset +139.7%
  7. stack-v3-train +132.3%
  8. OpenAlex +121.3%
  9. Berkeley-Function-Calling-Leaderboard +117.5%
  10. SAGE-10k +108.9%

By Category

  1. Other 73 · 65.9M
  2. Vision & Video 29 · 9.4M
  3. Text & Language Data 25 · 76.3M
  4. Robotics & Simulation 10 · 14.7M
  5. Question Answering & Reasoning 6 · 5.7M
  6. Tabular & Time-Series 4 · 436.9k
  7. Speech & Audio 3 · 2.1M

How we rank, and why you can trust it

  • Data source: the free, public Hugging Face Hub API (downloads, likes, license, and last-modified date for every dataset). Refreshed weekly.
  • Inclusion is automatic: a dataset must clear an adoption floor - at least 1,000 downloads in the last 30 days, or at least 50 likes - so the list stays credible rather than padded with unused uploads. We do not hand-pick winners.
  • Adoption is measured by all-time downloads (when available) and likes, log-scaled so a handful of enormous outliers do not flatten the rest of the ranking.
  • Growth is measured by the 30-day change in downloads, compared against our own weekly snapshots - shown once enough history has accrued.
  • Maintenance reflects how recently a dataset was updated - a dataset updated in the last 30 days scores highest, decaying the longer it has gone untouched.
  • Ranking is data-driven only. No dataset pays to be listed or ranked.

The Quality Score formula

The Quality Score (0-100) is a transparent weighted blend: 40% Adoption (downloads and likes, log-scaled), 20% Growth (30-day downloads trend), 25% Maintenance (how recently the dataset was updated), and 15% Trust (a stated license, an identified author, and not being access-gated).

Frequently asked questions

What are the best Hugging Face datasets?

The best Hugging Face datasets by our composite data score are listed in the leaderboard above, ranked on adoption, growth, and maintenance rather than opinion. Use the filters to narrow by task category.

Are Hugging Face datasets free to use?

Most datasets on the Hugging Face Hub are free and openly licensed, though terms vary by dataset. Some datasets are "gated" and require accepting terms before download. Each row shows the license when Hugging Face reports one.

How often is this list updated?

This directory refreshes weekly, pulling fresh downloads, likes, and maintenance data directly from the Hugging Face Hub API.

How many datasets are on the Hugging Face Hub?

The Hugging Face Hub hosts well over 200,000 datasets. This directory ranks the 150 that clear our adoption floor (at least 1,000 downloads in 30 days, or 50 likes), as of August 4, 2026.

See also: Hugging Face Models, ranked, MCP Servers directory, and WordPress AI Plugins directory.

Last updated: . Data from the Hugging Face Hub API. See the methodology and the machine-readable data.json.