Wan2.1

Alibaba Tongyi's video creation/editing model — R2V/V2V/MV2V all-rounder

★ 17,074 stars, verified 2026-09-30Category: Video / Digital Humans / RestorationGitHub repository ↗

Alibaba Tongyi's video creation and editing model, an R2V/V2V/MV2V all-rounder; Wan2.2 is already in the pool (category B) — this repo is the 2.1 generation.

AI video generation, digital humans, and video restoration.

Key features

Open suite of video foundation models: text-to-video, image-to-video, video editing
T2V-1.3B runs on 8.19GB VRAM; 5-second 480P video on RTX 4090 in about 4 minutes
First video model capable of generating both Chinese and English visual text
Wan-VAE encodes/decodes 1080P videos of any length while preserving temporal info
Integrated into Diffusers and ComfyUI; weights on Hugging Face and ModelScope

Good for

Open-source video generation on consumer-grade GPUs
Research and community extensions (video editing, motion control

Stack & tags

PythonPyTorchvideo-generationtext-to-videoimage-to-videoopen-weightsdiffuserscomfyuiwan

Similar open-source projects

Hyperframes

★ 55,468

Write HTML. Render video. Built for agents — HeyGen's open-source video framework

Video / Digital Humans / RestorationGitHub ↗

Hypit

★ 18,649

Clone any viral video with AI agents — swap the face, the words, the B-roll, ship 100 variants

Video / Digital Humans / RestorationGitHub ↗