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

Datasetmlfoundations/dcvlm-baseline-200bTop pick

DCVLM-Baseline multimodal corpus with 200B tokens

Cleaned mixture of 103.9 million image-text pairs in 6 languages in WebDataset format for VLM pre-training.

  • Image text → text
  • Visual question answering
  • English
  • Chinese
  • German
  • +9
77.1K 910K–100K rows
DatasetVoxel51/mvtec-ad

MVTec AD industrial defect dataset in FiftyOne format

Well-known benchmark for anomaly detection and manufacturing defects (5,354 images) adapted for FiftyOne.

  • Image classification
  • Image segmentation
  • English
76.7K 181K–10K rows
Datasetallenai/MolmoAct-Midtraining-MixtureTop pick

Multimodal data for robot manipulation training

Image and text mixture for the midtraining phase of the open MolmoAct robotics model by Allen Institute for AI.

  • Robotics
76K 111M–10M rows
DatasetAngais/pixel-art-atlas-1.4m

1.4m pixel art sprites with captions

Synthetic 16x16 game sprites with English captions for training text-to-image models and LoRA.

  • Text → image
  • English
74K 6701M–10M rows
DatasetFsoft-AIC/RobotDesign1MTop pick

Multimodal Robot Design Dataset

Millions of image-text pairs from research papers for robot design generation and analysis.

  • Image text → text
  • Text → image
73.6K 161M–10M rows
Datasetuoft-cs/cifar100

CIFAR-100: 60k Images Across 100 Classes

A 32×32 color image dataset for classification across 100 classes and 20 superclasses, with 50k train and 10k test.

  • Image classification
  • English
72.5K 7710K–100K rows
DatasetInternRobotics/InternData-A1Top pick

Robot manipulators: 630K trajectories

Hybrid synthetic-real trajectory dataset for training robotic manipulators.

  • Other
  • Robotics
  • English
71.9K 118over 1T rows🔒 on request
Datasetylecun/mnist

MNIST 28x28 handwritten digits

70k black-and-white 28x28 handwritten digit images for training and validating classifiers.

  • Image classification
  • English
70.1K 30210K–100K rows
Dataseteinrafh/hnm-fashion-recommendations-data

H&M transactions and product catalog for recommendations

Large array of anonymized purchases, customer profiles, and H&M clothing metadata for recommendation systems.

  • Tabular classification
  • Text classification
  • Image text → text
  • English
  • Indonesian
69.2K 510M–100M rows
Datasetmlfoundations/dcvlm-balanced-200bTop pick

Balanced multimodal DCVLM corpus of 200B tokens

Cleaned dataset of 112M image-text pairs in 6 languages in WebDataset format for pretraining multimodal models.

  • Image text → text
  • Visual question answering
  • English
  • Chinese
  • German
  • +9
67.7K 210K–100K rows
DatasetUnipatAI/RoadmapBench

Coding agent benchmark on repository upgrades

115 complex software development tasks from real open-source project updates across five programming languages.

  • Text generation
  • English
63.2K 12up to 1K rows
Datasetdeepghs/generic_character_skins

Anime, manga, and game character images from Zerochan

Thousands of character artworks across over 2,000 folders for classification and caption generation.

  • Image classification
  • Image → text
  • Other
  • English
  • Japanese
  • Chinese
62.6K 2
Datasetnvidia/USDCraft-10K

NVIDIA OpenUSD simulation 3D assets

1,000 to 10,000 movable and rigid 3D objects with physics and joints for robot training.

  • Robotics
59.7K 14510K–100K rows