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

DatasetSpatialVID/SpatialVIDTop pick

Large-scale video dataset with spatial annotations

1 to 10 million entries with spatial annotations for video and 3D generation tasks.

  • Text → video
  • Text → 3d
  • Image → 3d
  • English
14.2K 491M–10M rows🔒 on request
DatasetDAGroup-PKU/RoVid-XTop pick

Robotic videos for embodied generation

4M robotic clips with RGB, depth, and optical flow for image-to-video.

  • Image → video
  • English
13.2K 74over 1T rows
Datasetomegalabsinc/omega-multimodal

Multimodal video dataset from the Bittensor network

A massive video corpus with text descriptions and tabular metadata from the decentralized OMEGA Labs network.

  • Video text → text
  • Video classification
  • Image classification
11.3K 61
DatasetRapidata/text-to-3d-with-llm-code

LLM 3D code generation benchmark with human ratings

44,482 examples and over 780,000 human ratings to test LLM capability in creating 3D objects via code.

  • Text → video
  • English
7.4K 5110K–100K rows
DatasetFelixYuan-YF/SpatialVID-HQ

Large video dataset with spatial annotations

A large-scale dataset featuring videos and spatial annotations for 3D and video generation tasks.

  • Text → video
  • Text → 3d
  • Image → 3d
  • English
5.6K 35100K–1M rows🔒 on request
DatasetNU-World-Model-Embodied-AI/phyground

Physics benchmark in generated video

250 prompts with initial frames and labels for 13 laws of physics to test the physical validity of video generation models.

  • Text → video
  • Image → video
  • English
3.9K 48up to 1K rows
DatasetAlayaLab/WildWorld

Video dataset for ARPG genre game world models

Large video dataset with actions and states for training generative game world models.

  • Text → video
  • Image → video
  • Video → video
  • English
3.4K 15🔒 on request
Datasetjovianzm/Pexels-400k

400k Pexels Videos with Captions and Metadata

400,476 videos with previews, view counts, explicit flags, and captions for video-text models.

  • Image → text
  • Text → image
  • Text → video
  • English
2.4K 70100K–1M rows
DatasetSoul-AILab/VividHeadTop pick

330k talking head clips in 15+ languages

782 hours of head-cropped speech videos: 330k clips, 15+ languages, varying ages.

  • Image → video
1.8K 69100K–1M rows
DatasetHKUSTAudio/Talker-T2AV-Data

Talker-T2AV talking person audio-video dataset

A cleaned multimodal corpus for joint speech and video generation of talking persons in Chinese and English.

  • Text → video
  • Text → speech
  • Chinese
  • English
1.8K 7100K–1M rows
Datasetsaiyan-world/Goku-MovieGenBench

Goku-MovieGenBench video data

Video data for training and evaluating the Goku text-to-video generative model.

  • Text → video
1.2K 2251K–10K rows
DatasetRapidata/text-2-video-human-preferences-moonvalley-marey

Human preference evaluations for Marey video generation

1,313 examples with 74.8 thousand human ratings for videos generated by the Marey model.

  • Video classification
  • Text → video
  • Text classification
  • English
1.2K 91K–10K rows
DatasetTencentARC/MiraData

Long videos with structured descriptions

A large Tencent dataset with long video clips and detailed text descriptions for training video generation.

  • Image → video
  • Text → image
  • Text → video
  • English
1.1K 43100K–1M rows
DatasetAPRIL-AIGC/Soul-Bench

Multimodal benchmark for avatar animation

Dataset of up to 1 thousand examples for evaluating image-to-video generation and digital human animation.

  • Image → video
730 56up to 1K rows
Datasetccmoony/PhyCoVE

Physically grounded video editing benchmark

508 tasks for altering physical object properties in video with reference clips and English/Chinese instructions.

  • Video → video
  • English
  • Chinese
615 2up to 1K rows
Datasetego2act/ego2act-bench

Ego2act benchmark for video generation

A benchmark for evaluating goal-directed manipulations in video generation using egocentric frames.

  • Image → video
  • Video classification
  • English
533 4up to 1K rows