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

Datasetj-shen3/MAVOS-DD

MAVOS-DD multilingual deepfake benchmark

Multilingual open audio-video deepfake benchmark with train and validation splits.

  • Video classification
  • Arabic
  • Romanian
  • English
  • +4
119.7K 010K–100K rows
Datasetcadene/droidTop pick

Large DROID robotics dataset in LeRobot format

Over 27 million video frames and 92 thousand robot manipulation episodes with task text descriptions.

  • Robotics
  • English
114.2K 3810M–100M rows
Datasetallenai/MolmoAct2-BimanualYAM-DatasetTop pick

Bimanual robotics demonstrations for MolmoAct2

Over 720 hours of dual-arm robot manipulation recordings with text annotations for training robotic models.

  • Robotics
113.3K 1810M–100M rows
Datasetbehavior-1k/2025-challenge-demosTop pick

Demos for R1Pro robot (90K videos)

A massive corpus of 10,000 episodes and 90,000 videos for robot training.

  • Robotics
112.3K 44
Datasetmarkov-ai/cad-1000-hours

CAD Tasks with Video and Reference Files

509 end-to-end CAD tasks across seven programs: video, input and output files, and evaluation criteria.

    105.4K 612
    Datasetnkp37/OpenVid-1MTop pick

    OpenVid-1M text-captioned video dataset

    Large collection of video clips with English captions for training video generation models.

    • Text → video
    • Image → video
    • English
    95.2K 3001M–10M rows
    Datasetzekaiwang/trex_datasetTop pick

    Bimanual tactile manipulation data for robots

    Teleoperation recordings of a robot with two five-fingered hands for training robotics models.

    • Robotics
    • English
    90.1K 361M–10M rows
    Datasetyaak-ai/L2DTop pick

    Largest multimodal autonomous driving dataset

    Over 5,000 hours of driving across 30 German cities with 6-camera video, telemetry, and environment labels.

    • Robotics
    90K 6110M–100M rows
    DatasetRealmanRobot/RealSource-WorldTop pick

    Dual-arm humanoid robot manipulation dataset

    Over 14 million frames and 11,400 RS-02 robot control episodes for real-world tasks.

    • Robotics
    • English
    • Chinese
    89.9K 3210K–100K rows🔒 on request
    Datasetlmms-lab/EgoLife

    Chinese video-text dataset

    A collection of videos and texts for video-text-to-text tasks, currently undergoing cleaning.

    • Video text → text
    • Chinese
    85.6K 3410K–100K rows
    DatasetHRDexDB/HRDexDBTop pick

    4D human and robot object grasping dataset

    Multimodal 3D and video dataset from Seoul National University for learning dexterous grasping.

      85.5K 2510K–100K rows
      Datasettumor-vqa/DeepTumorVQA_2.0Top pick

      Medical Visual Question Answering Benchmark

      3D CT abdominal benchmark for Visual Question Answering with 438K QA pairs and clinical subtypes.

      • Visual question answering
      • Image classification
      • English
      85.1K 810K–100K rows
      DatasetTacVerse/TacVerseTop pick

      Tactile-visual dataset for dual-arm robotics

      Large collection of synchronized video and tactile sensor data for training robot manipulation.

      • Robotics
      84.1K 26100M–1B rows🔒 on request
      DatasetHuggingFaceVLA/community_dataset_v3Top pick

      Cross-platform robot dataset from Hugging Face

      Large open video dataset for training Vision-Language-Action models across 46 robot types.

      • Robotics
      • English
      81K 2510K–100K rows
      Datasetcadene/agibot_alpha_v30Top pick

      AgiBot A2D robot trajectory logs in LeRobot

      Large robotics dataset with video, tabular telemetry, and text descriptions for 30 tasks.

      • Robotics
      76K 1010M–100M rows
      Datasetfpvlabs/stereo-550

      First-person stereo video of hand manipulations

      Calibrated stereo RGB videos of everyday hand actions for robotics and 3D vision.

      • Robotics
      • Video classification
      • Image → text
      • English
      75.2K 41K–10K rows🔒 on request
      Datasetfpvlabs/stera-10mTop pick

      Stera-10m: egocentric multimodal dataset

      200 hours of synchronized first-person recordings with RGB, LiDAR depth, audio, and text for embodied AI and robotics.

      • Robotics
      • Video classification
      • Image → text
      • English
      72.2K 881M–10M rows🔒 on request
      Datasetnvidia/PhysicalAI-WorldModel-Synthetic-Autonomous-Driving-ScenariosTop pick

      NVIDIA autonomous driving synthetic camera videos

      Synthetic panoramic traffic video clips with VLM descriptions for training world models and autopilot.

      • English
      71.4K 44100K–1M rows
      DatasetBitRobot/HIW-500-LeRobot

      Robotics dataset in LeRobot format

      A large array of English video, tabular, and time-series data for training robot control systems.

      • Robotics
      • English
      70.7K 2510M–100M rows