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

DatasetIDEAS-Lab-Northwestern/ManiGuard-Bench

Home robotics safety benchmark

1,000 simulations to verify how accurately and safely robots perform household tasks.

  • Robotics
45.8K 21K–10K rows
DatasetIcey444/retdata

Million-Scale Reasoning Image Retrieval Dataset

Over 1.7 million source photo and instruction to target photo pairs for multimodal reasoning retrieval training.

  • Image → image
45.7K 01M–10M rows🔒 on request
DatasetVoxel51/MPII_Human_Pose_Dataset

MPII Human Pose annotations on images

About 25K YouTube frames with joint annotations for over 40K people for pose and action estimation.

  • Image classification
  • Object detection
  • English
45.4K 1810K–100K rows
Datasetnewniuuu/OpenFly-traj-converted

Converted OpenFly Trajectories

Converted trajectory subsets from the OpenFly dataset for robotics applications.

  • Robotics
  • English
45.3K 0100K–1M rows
Datasetvankey/RealText-V2Top pick

Multilingual benchmark for document forgery detection

Over 20K document images across six languages for forgery detection and fake text analysis.

  • Image segmentation
  • Text classification
  • Visual question answering
  • English
  • Chinese
  • Arabic
  • +3
44.7K 810K–100K rows
Datasetonboard-coop/STARCOP-fast-products

Hyperspectral Satellite Images for Methane Search

Satellite data for methane emission segmentation, prepared for lightweight onboard neural networks.

  • Image segmentation
44.5K 11K–10K rows
Datasetvector-institute/open-pmc-18mTop pick

Medical Images and Captions from PubMed Central

Large-scale English corpus of image-text pairs extracted from open-access PubMed Central medical articles.

  • English
44K 1010M–100M rows
DatasetRekaAI/RekaDaily-10k-processedTop pick

First-person daily videos with captions and Q&A

Millions of short daily first-person videos with detailed captions and Q&A from Reka AI.

  • Video text → text
  • Visual question answering
  • Video classification
  • English
44K 121M–10M rows
DatasetE4DRR/gik-ecmwf-par

ECMWF weather forecast virtual parquet indexes

Lightweight parquet files with links to ECMWF IFS weather forecasts on AWS S3 for analysis without downloading gigabytes.

    43K 4up to 1K rows
    Datasetcodepawl/simready-glb

    SimReady 3D models in GLB format

    2,595 web-ready .glb models from nvidia/simready-assets for three.js, Babylon.js, and other glTF 2.0 viewers.

      42.9K 1051K–10K rows
      Datasetcommon-canvas/commoncatalog-cc-by-nc-sa

      Large 4K image corpus with synthetic captions

      Tens of millions of high-resolution Flickr images with English captions for image generation.

      • Text → image
      • English
      41.8K 510M–100M rows
      Datasetdata-is-better-together/open-image-preferences-v1

      Paired comparisons of generated images by prompt

      English prompt dataset with image pairs for evaluating text-to-image preferences.

      • Text → image
      • Image → text
      • English
      40.8K 341K–10K rows
      Datasetgagandeepreehal/indus-ad-monocam

      Monocular car camera recordings on Indian roads

      96 single-camera driving videos and over 105 thousand frames with depth and semantic segmentation annotations.

      • Robotics
      • Depth estimation
      • Image segmentation
      • English
      40.6K 1100K–1M rows🔒 on request
      DatasetMMMU/MMMU_Pro

      MMMU-Pro multimodal benchmark

      Multidisciplinary visual and textual questions for multimodal understanding evaluation.

      • Question answering
      • Visual question answering
      • Multiple choice
      • English
      40.4K 1031K–10K rows
      DatasetSandflow/sand3d-articraft

      Parametric 3D objects as Python code

      9,209 everyday items, mechanisms, and CAD parts provided as sand3d code, OpenUSD (.usda) files, and renders.

      • Text → 3d
      • Image → 3d
      40.4K 81K–10K rows
      DatasetXiangpengYang/VideoGrain-dataset

      Video editing dataset for VideoGrain

      English image and data collection from the VideoGrain paper (ICLR 2025) for multi-level video editing tasks.

      • Text → video
      • Text → image
      • English
      40.3K 11K–10K rows