SkyReels V1

HunyuanVideo-based human-centric video model that fits one 24 GB GPU.

Install Repository Compare

Fit.

Catalog snapshot

Platforms: Linux, Macos, Windows
Operating systems: Linux, macOS, Windows
Ram floor: 32 GB
Gpu: Required · 24 GB VRAM
Docker: No
Setup: Advanced
Deploy as: Local
Commercial use: Check the license

Custom license — check terms before commercial use.

Project.

Fetched Sep 12, 2026¹

License: Other
Stars: 2.7k¹
Forks: 311¹
Open issues: 53¹
Language: Python¹
Last commit: Mar 10, 2025¹
Tags: videohuman-centricquantized

About.

SkyReels-V1 tunes HunyuanVideo for human-centric shots — faces, hands, and body motion that holds together. INT4 quantization fits a single 24 GB card. Custom model license; the repo has been quiet since early 2025.

human-centric motion · SkyReels-V2 infinite · ComfyUI workflows · INT4 quantization

Alternatives.

Nearby.

All Video AI

WWan2.1Alibaba’s open text-to-video and image-to-video stack.Compare
OOpen-SoraApache-2.0 research stack for training and running video diffusion on 24 GB cards.Compare
CCogVideoXTHUDM open video diffusion with strong prompt following.Compare
FFramePackLong local image-to-video with a low VRAM floor.Compare
LLTX-VideoLightricks open video model aimed at faster local clips.Compare