Skip to page content
SAVRN
Search Plan a project

Dataset · Text to video

anime-ai-training-dataset

by H tertrxcuvuvjvj/anime-ai-training-dataset

Curated high-quality anime sakuga animation dataset tailored for AI video model training (AnimateDiff, Wan 2.1, HunyuanVideo, CogVideoX, SVD, LTX-Video, etc.). -: Sword fights, hand-to-hand combat, rapid movement.

Rows—
Configurations—
Size419.2 MB
Licensemit
AccessPublicly accessible
Monthly Downloads—

Dataset Card

By H, published under mit, revision e8f05b6498bc.

Anime AI Training Dataset (100+ High Quality Clips, 5-8s)

Curated high-quality anime sakuga animation dataset tailored for AI video model training (AnimateDiff, Wan 2.1, HunyuanVideo, CogVideoX, SVD, LTX-Video, etc.).

Features:

  • Clip Durations: Uniform 5 to 8 seconds per clip (ideal for temporal consistency and frame batching).
  • Format: MP4 (lossless cut, clean audio/video streams).
  • Categories:
  • : Sword fights, hand-to-hand combat, rapid movement.
  • : Running, acrobatics, jumps, dynamic perspective transitions.
  • : Subtle character acting, expressions, hair and cloth physics.
  • : Fire, water, wind, debris, lightning, energy beams.

Direct Download:

Download the complete dataset package directly using the link below: - anime_ai_training_100_clips_5to8s.zip (~400 MB)

Details

Repository
tertrxcuvuvjvj/anime-ai-training-dataset
Publisher
H
Task category
Text to video
Tags
anime, sakuga, animation
Size category
n<1K
Languages
Not stated by the source
Revision
e8f05b6498bc4b10ee3e4bd27ca97b20ccb88f9f
Last updated
2026-09-23

Files

3 files, 419.2 MB in total.

Data1 file · 419.2 MB
Documentation1 file · 1.2 KB
Repository1 file · 2.5 KB
Every file
FileTypeSizeSHA-256
anime_ai_training_100_clips_5to8s.zipData419.2 MB2038eea19102
README.mdDocumentation1.2 KB—
.gitattributesRepository2.5 KB—

License and Download

License
mit
Access
No access gate
Download from H

Released by H through its official repository on Hugging Face. Read the license.