In this repository, we present Wan2.1, a comprehensive and open suite of video foundation models that pushes the boundaries of video generation. Wan2.1 offers these key features:

👍 SOTA Performance: Wan2.1 consistently outperforms existing open-source models and state-of-the-art commercial solutions across multiple benchmarks.
👍 Supports Consumer-grade GPUs: The T2V-1.3B model requires only 8.19 GB VRAM, making it compatible with almost all consumer-grade GPUs. It can generate a 5-second 480P video on an RTX 4090 in about 4 minutes (without optimization techniques like quantization). Its performance is even comparable to some closed-source models.
👍 Multiple Tasks: Wan2.1 excels in Text-to-Video, Image-to-Video, Video Editing, Text-to-Image, and Video-to-Audio, advancing the field of video generation.
👍 Visual Text Generation: Wan2.1 is the first video model capable of generating both Chinese and English text, featuring robust text generation that enhances its practical applications.
👍 Powerful Video VAE: Wan-VAE delivers exceptional efficiency and performance, encoding and decoding 1080P videos of any length while preserving temporal information, making it an ideal foundation for video and image generation.

This repository features our T2V-14B model, which establishes a new SOTA performance benchmark among both open-source and closed-source models. It demonstrates exceptional capabilities in generating high-quality visuals with significant motion dynamics. It is also the only video model capable of producing both Chinese and English text and supports video generation at both 480P and 720P resolutions.

METAFILM_Ai

License:

Apache 2.0

通义实验室 WAN 2.1 Model Zoo

Institute for Intelligent Computing专注于各领域大模型技术研发与创新应用。实验室研究方向涵盖自然语言处理、多模态、视觉AIGC、语音等多个领域。我们并积极推进研究成果的产业化落地。实验室同时积极参与开源社区建设，全方位拥抱开源社区，共同探索AI模型的开源开放。

Developer / Models Name / Kijai`s ComfyUI Model

Wan Team/1.3B text-to-video included / 文生视频1.3B 已收录

Wan Team/14B text-to-video included / 文生视频14B 已收录

Wan Team/14B image-to-video 480P included / 图生视频14B 已收录

Wan Team/14B image-to-video 720P included / 图生视频14B 已收录

Wan Team/14B first-last-frame-to-video 720P included / 视频首尾帧 已收录

DiffSynth-Studio Team/1.3B aesthetics LoRA 美学蒸馏-通义万相2.1-1.3B-LoRA-v1

DiffSynth-Studio Team/1.3B Highres-fix LoRA 高分辨率修复-通义万相2.1-1.3B-LoRA-v1

DiffSynth-Studio Team/1.3B ExVideo LoRA 长度扩展-通义万相2.1-1.3B-LoRA-v1

DiffSynth-Studio Team/1.3B Speed Control adapter 速度控制-通义万相2.1-1.3B-适配器-v1

PAI Team/ WAN2.1 Fun 1.3B InP 支持首尾帧 / Kijai/WanVideo_comfy

PAI Team/ WAN2.1 Fun 14B InP 支持首尾帧 / Kijai/WanVideo_comfy

PAI Team/ WAN2.1 Fun 1.3B Control 控制器 / Kijai/WanVideo_comfy

PAI Team/ WAN2.1 Fun 14B Control 控制器 / Kijai/WanVideo_comfy

IIC Team/ VACE-通义万相2.1-1.3B-Preview / Kijai/WanVideo_comfy

REDCraft AIGC / WAN2.1 720P NSFW Unlocked FT / Coming soon

WAN Comfy-Org/Wan_2.1_ComfyUI_repackaged

【例图页面蓝色Nodes或下载webp文件-可复现视频工作流】

Gallery sample images/videos (WEBP format) including the ComfyUI native workflow

This is a concise and clear GGUF model loading and tiled sampling workflow：

Wan 2.1 Low vram Comfy UI Workflow (GGUF) 4gb Vram - v1.1 | Wan Video Workflows | Civitai

节点：（或使用 comfyui manager 安装自定义节点）

https://github.com/city96/ComfyUI-GGUF

https://github.com/kijai/ComfyUI-WanVideoWrapper

https://github.com/BlenderNeko/ComfyUI_TiledKSampler

* 注意需要更新到最新版本的 comfyui-KJNodes GitHub - kijai/ComfyUI-KJNodes: Various custom nodes for ComfyUI update to the latest version of Comfyui KJNodes

Kijai ComfyUI wrapper nodes for WanVideo

WORK IN PROGRESS

@kijaidesign 's works

Huggingface - Kijai/WanVideo_comfy

GitHub - kijai/ComfyUI-WanVideoWrapper

主图视频来自 AiWood

https://www.bilibili.com/video/BV1TKP3eVEue

Text encoders to ComfyUI/models/text_encoders

Transformer to ComfyUI/models/diffusion_models

Vae to ComfyUI/models/vae

Right now I have only ran the I2V model succesfully.

Can't get frame counts under 81 to work, this was 512x512x81

~16GB used with 20/40 blocks offloaded

DiffSynth-Studio Inference GUI

Wan-Video LoRA & Finetune training.

DiffSynth-Studio/examples/wanvideo at main · modelscope/DiffSynth-Studio · GitHub

Wan: Open and Advanced Large-Scale Video Generative Models

通义万相Wan2.1视频模型开源！视频生成模型新标杆，支持中文字效+高质量视频生成

In this repository, we present Wan2.1, a comprehensive and open suite of video foundation models that pushes the boundaries of video generation. Wan2.1 offers these key features:

👍 SOTA Performance: Wan2.1 consistently outperforms existing open-source models and state-of-the-art commercial solutions across multiple benchmarks.
👍 Supports Consumer-grade GPUs: The T2V-1.3B model requires only 8.19 GB VRAM, making it compatible with almost all consumer-grade GPUs. It can generate a 5-second 480P video on an RTX 4090 in about 4 minutes (without optimization techniques like quantization). Its performance is even comparable to some closed-source models.
👍 Multiple Tasks: Wan2.1 excels in Text-to-Video, Image-to-Video, Video Editing, Text-to-Image, and Video-to-Audio, advancing the field of video generation.
👍 Visual Text Generation: Wan2.1 is the first video model capable of generating both Chinese and English text, featuring robust text generation that enhances its practical applications.
👍 Powerful Video VAE: Wan-VAE delivers exceptional efficiency and performance, encoding and decoding 1080P videos of any length while preserving temporal information, making it an ideal foundation for video and image generation.