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

DatasetChartGalaxy/ChartGalaxy

Infographic Charts for VQA

Infographic charts with text annotations for visual understanding and generation tasks.

  • Visual question answering
  • English
18.9K 891K–10K rows
DatasetHuggingFaceM4/WebSight

WebSight: screenshots and website HTML/CSS

A large synthetic dataset of screenshot-to-HTML/CSS-code pairs for image-based UI generation.

  • English
18.9K 4121M–10M rows
Datasetmrmrx/CADS-datasetTop pick

Full-body 3D CT scans with 167 organ annotations

Over 22 thousand volumetric head-to-knee CT scans with full segmentation of 167 anatomical structures.

  • Image segmentation
18.1K 7010K–100K rows🔒 on request
Datasetzalando-datasets/fashion_mnist

Fashion-MNIST: 70k clothing images

60k train and 10k test 28x28 grayscale clothing images with 10 classes for classification.

  • Image classification
  • English
17.8K 6910K–100K rows
Datasetmrlbenchmarks/global-piqa-nonparallelTop pick

Common sense benchmark for 136 language variants

Multiple-choice questions to test understanding of everyday logic and cultural features in rare languages.

  • Question answering
  • Multiple choice
  • acm
  • acq
  • Tunisian Arabic
  • +9
17.6K 4310K–100K rows
DatasetOleehyO/latex-formulas-80M

Large image corpus with LaTeX formulas

Tens of millions of pairs of images and LaTeX markup for training formula recognition models.

    17.6K 2910M–100M rows
    Datasettasksource/tasksource-jev-typed-decisions

    Typed decisions from hundreds of Tasksource tasks

    Millions of classification, choice, rating, and visual reasoning examples for multi-task learning.

    • Zero shot classification
    • Text classification
    • Question answering
    • English
    • Multilingual
    17.5K 341M–10M rows
    DatasetMajor-TOM/Core-S2L2A

    Global Sentinel-2 Level 2A imagery

    Over 2.2 million global Sentinel-2 Level 2A patches of 1068x1068 pixels for geospatial tasks.

      16.6K 681M–10M rows
      DatasetUCSC-VLAA/GPT-Image-Edit-1.5MTop pick

      One million triplets for image editing

      Source image, instruction, and edited image pairs for training and evaluating image-to-image models.

      • Image → image
      • English
      16.5K 971M–10M rows
      Datasethongzhuyi/VDEBench

      Document and Table Pointwise Editing Benchmark

      A set of 3,506 examples for evaluating models that edit text and tables on document scans via instructions.

      • Image → image
      • English
      • Chinese
      16.4K 621K–10K rows
      Datasetgoogle/RSRCCTop pick

      Google Remote Sensing Change Captioning Benchmark

      A Google Research dataset with multi-temporal aerospace image pairs and questions about terrain changes.

      • Visual question answering
      • Image text → text
      • Multiple choice
      • English
      16.2K 41100K–1M rows
      Datasetccore/ThreeJS-3D-ScenesTop pick

      Three.js 3D scenes with prompts and code

      4,146 examples of Three.js 3D scenes with prompts, thinking traces, and ready-to-use HTML/JavaScript code.

      • Text generation
      • English
      16.1K 1821K–10K rows
      DatasetMathLLMs/MathVision

      MATH-Vision benchmark for multimodal mathematics

      Visual math problems with images and text for evaluating VLMs.

      • Question answering
      • Multiple choice
      • Visual question answering
      • English
      15.7K 1841K–10K rows
      DatasetMAmmoTH-VL/MAmmoTH-VL-Instruct-12M

      12M visual instructions for multimodal models

      Image-instruction pairs for training visual assistants in English.

      • Visual question answering
      • Question answering
      • English
      15.4K 6710M–100M rows
      DatasetLucasFang/FLUX-Reason-6MTop pick

      6M images for text-to-image reasoning

      6 million images generated by FLUX.1-dev and 20 million English-Chinese descriptions for training generative models.

        15.3K 1221M–10M rows
        Datasetosunlp/MagicBrush

        Instructions for image editing

        10K triplets of source image, instruction, and result for training editing.

        • Text → image
        • Image → image
        • English
        15.2K 11710K–100K rows
        DatasetOpen-Bee/Honey-Data-15M

        Bee multimodal image-text corpus

        15M English image-text pairs for training and evaluating open multimodal models.

        • Image text → text
        • English
        15K 12110M–100M rows
        DatasetQwen/Qwen-Image-BenchTop pick

        Qwen Image Generation Benchmark

        An English and Chinese image-text dataset for deep evaluation of Text-to-Image models.

        • Image → text
        • English
        • Chinese
        15K 531K–10K rows
        DatasetSandflow/sand3d-simready

        Procedural 3D objects as Python code and OpenUSD

        2,334 parametric 3D models and mechanisms with sand3d code, OpenUSD files, and renders for 3D generation.

        • Text → 3d
        • Image → 3d
        14.7K 151K–10K rows