Skip to content

Datasets

Hugging Face datasets with plain-language descriptions: what's inside and what models to train on them. 4,788 in the catalog so far, with new ones added every hour.

Data type

For which models

Find a dataset for your goal

Answered by Gemini 4 · the chat is kept only in this tab

Datasetthetrademarkk/india-index-options-1m

Indian index and option minute quotes

A massive archive of NSE/BSE spot index and option minute candles for 2021–2026 in parquet format.

  • English
22.3K 8100M–1B rows
Datasetbens-bots/marrs-global-coral-reef-soundscapes

Coral reef soundscapes in parquet format

Millions of audio recordings from coral reef restoration sites with sound type annotations for bioacoustics.

  • Audio classification
  • English
22.3K 11M–10M rows
DatasetMcGill-NLP/agent-reward-bench

Web agent auto-evaluation benchmark

A set of 1–10k English web agent trajectories for evaluating automated evaluator quality.

  • Robotics
  • English
22.3K 71K–10K rows
DatasetMCG-NJU/VideoChat3-LV116k

Long video analysis instructions (VideoChat3)

An English video-text instruction dataset for training models to understand long videos.

  • Video text → text
  • English
22.2K 151K–10K rows
DatasetLocalLLaMA/typed-decisions

Typed probabilistic decisions benchmark

Benchmark: the model receives an unstructured state and answers five typed questions with probabilities.

  • Text classification
  • English
22.2K 2781K–10K rows
DatasetKodCode/KodCode-Light-RL-10K

KodCode-Light synthetic coding dataset

A large synthetic dataset featuring verified solutions and tests for programming tasks.

  • Question answering
  • English
22.2K 1010K–100K rows
Datasetnebius/SWE-agent-trajectoriesTop pick

80k SWE-agent trajectories

Programmatic agent trajectories solving GitHub issues to train coding agents.

    22.2K 10710K–100K rows
    DatasetMedOtter/4D-Lung

    4D lung CT segmentation dataset

    Respiratory-synchronized CT scan dataset of lung cancer patients for segmentation.

    • Image segmentation
    22.1K 2up to 1K rows
    Datasetnvidia/Nemotron-SFT-Science-v2Top pick

    Nemotron Science SFT dataset

    A large dataset of scientific reasoning for physics, chemistry, and biology featuring synthetic and real tasks.

    • Text generation
    • English
    22K 211M–10M rows
    Datasetmaelic/VG150-coco-format

    Visual Genome 150 (VG150) in COCO-JSON

    Popular scene graph generation benchmark with 150 object classes and 50 relation types in standard COCO format.

    • Object detection
    • English
    22K 0100K–1M rows
    DatasetNuBerea/reformationTop pick

    Reformation reception corpus with OCR

    A corpus of Reformation-era texts featuring page-grain OCR, annotation, and lemmas in multiple languages.

    • Embeddings
    • Latin
    • English
    • German
    • +1
    22K 110M–100M rows🔒 on request
    Datasetewinchell07/so101-smolvla-finetuned-250-eval-20261003

    SmolVLA robotics evaluation logs

    SmolVLA robotics model evaluation containing 250 completed camera trials.

    • Robotics
    21.9K 0up to 1K rows
    Datasetmorgan/CADExpert_fixed

    CADExpert dataset of CAD tasks

    14,935 unique CAD tasks with images and three expert answers each.

    • Image → text
    • English
    • Chinese
    21.9K 110K–100K rows
    DatasetHuggingFaceFW/finewiki

    Wikipedia in 325 languages: cleaned finewiki

    Updated Wikipedia from August 2025 HTML dumps: 325 languages, clean articles and metadata.

    • Text generation
    21.9K 34610M–100M rows
    DatasetRidheshBhati/Complete_Data_Source_100K_HOURS

    100,000-hour multilingual speech corpus

    A collection of audio recordings and texts in 6 languages, including English, Hindi, Bengali, and Tamil.

    • Assamese
    • Bangla
    • English
    • +9
    21.9K 41M–10M rows
    Datasetnvidia/PhysicalAI-Human-1M-100KTop pick

    Egocentric video and hand poses for physical AI

    An open portion of NVIDIA's Human-1M collection: first-person videos, head and hand poses with text descriptions.

    • English
    21.9K 129up to 1K rows
    DatasetHothan/OlympiadBench

    Multimodal benchmark of Olympiad scientific problems

    Complex Olympiad-level problems with images in English and Chinese for evaluating model scientific reasoning.

    • Question answering
    • Visual question answering
    • Chinese
    • English
    21.8K 521K–10K rows
    DatasetPKU-Alignment/BeaverTails

    QA pairs with harm annotations

    100K+ human-annotated QA pairs across 14 harm categories for AI safety.

    • Text classification
    • English
    21.8K 119100K–1M rows