diff --git a/README.md b/README.md
index 66b9fb3e9..43962ef52 100644
--- a/README.md
+++ b/README.md
@@ -36,6 +36,8 @@ We believe that a well-developed open-source code framework can lower the thresh
 
 > Currently, the development personnel of this project are limited, with most of the work handled by [Artiprocher](https://github.com/Artiprocher) and [mi804](https://github.com/mi804). Therefore, the progress of new feature development will be relatively slow, and the speed of responding to and resolving issues is limited. We apologize for this and ask developers to understand.
 
+- **August 7, 2026** We add support for Wan-Animate-2 in the Wan series. Given a reference image and a driving video, it makes the reference character perform the motions in the driving video, generating high-quality character animation, with both standard and distilled variants. For details, please refer to the [documentation](/docs/en/Model_Details/Wan.md) and [example code](/examples/wanvideo/).
+
 - **August 3, 2026** MiniMax-H3 open-sourced, welcome a new member to the video model family! Support includes text-to-video-audio generation, keyframe-guided generation, reference-driven generation, low VRAM inference, and NF4-quantized inference. For details, please refer to the [documentation](/docs/en/Model_Details/MiniMax-H3.md) and [example code](/examples/minimax_h3/).
 
 - **July 28, 2026** LingBot-Video open-sourced, welcome a new member to the video model family! This release includes two variants, Dense-1.3B and MoE-30B-A3B (30B total parameters, ~3B active per token), both supporting text-to-video, image-to-video and text-to-image generation, low VRAM inference, and LoRA / full training capabilities. For details, please refer to the [documentation](/docs/en/Model_Details/LingBot-Video.md) and [example code](/examples/lingbot_video/). Huge thanks to [NancyFyong](https://github.com/NancyFyong) for contributing the integration of this model!
@@ -1413,41 +1415,43 @@ Example code for Wan is available at: [/examples/wanvideo/](/examples/wanvideo/)
 
 | Model ID | Extra Inputs | Inference | Low VRAM Inference | Full Training | Validation After Full Training | LoRA Training | Validation After LoRA Training |
 |-|-|-|-|-|-|-|-|
-|[Wan-AI/Wan2.1-T2V-1.3B](https://modelscope.cn/models/Wan-AI/Wan2.1-T2V-1.3B)||[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.1-T2V-1.3B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.1-T2V-1.3B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.1-T2V-1.3B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.1-T2V-1.3B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.1-T2V-1.3B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.1-T2V-1.3B.py)|
-|[Wan-AI/Wan2.1-T2V-14B](https://modelscope.cn/models/Wan-AI/Wan2.1-T2V-14B)||[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.1-T2V-14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.1-T2V-14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.1-T2V-14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.1-T2V-14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.1-T2V-14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.1-T2V-14B.py)|
-|[Wan-AI/Wan2.1-I2V-14B-480P](https://modelscope.cn/models/Wan-AI/Wan2.1-I2V-14B-480P)|`input_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.1-I2V-14B-480P.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.1-I2V-14B-480P.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.1-I2V-14B-480P.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.1-I2V-14B-480P.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.1-I2V-14B-480P.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.1-I2V-14B-480P.py)|
-|[Wan-AI/Wan2.1-I2V-14B-720P](https://modelscope.cn/models/Wan-AI/Wan2.1-I2V-14B-720P)|`input_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.1-I2V-14B-720P.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.1-I2V-14B-720P.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.1-I2V-14B-720P.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.1-I2V-14B-720P.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.1-I2V-14B-720P.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.1-I2V-14B-720P.py)|
-|[Wan-AI/Wan2.1-FLF2V-14B-720P](https://modelscope.cn/models/Wan-AI/Wan2.1-FLF2V-14B-720P)|`input_image`, `end_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.1-FLF2V-14B-720P.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.1-FLF2V-14B-720P.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.1-FLF2V-14B-720P.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.1-FLF2V-14B-720P.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.1-FLF2V-14B-720P.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.1-FLF2V-14B-720P.py)|
-|[iic/VACE-Wan2.1-1.3B-Preview](https://modelscope.cn/models/iic/VACE-Wan2.1-1.3B-Preview)|`vace_control_video`, `vace_reference_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.1-VACE-1.3B-Preview.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.1-VACE-1.3B-Preview.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.1-VACE-1.3B-Preview.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.1-VACE-1.3B-Preview.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.1-VACE-1.3B-Preview.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.1-VACE-1.3B-Preview.py)|
-|[Wan-AI/Wan2.1-VACE-1.3B](https://modelscope.cn/models/Wan-AI/Wan2.1-VACE-1.3B)|`vace_control_video`, `vace_reference_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.1-VACE-1.3B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.1-VACE-1.3B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.1-VACE-1.3B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.1-VACE-1.3B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.1-VACE-1.3B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.1-VACE-1.3B.py)|
-|[Wan-AI/Wan2.1-VACE-14B](https://modelscope.cn/models/Wan-AI/Wan2.1-VACE-14B)|`vace_control_video`, `vace_reference_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.1-VACE-14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.1-VACE-14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.1-VACE-14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.1-VACE-14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.1-VACE-14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.1-VACE-14B.py)|
-|[PAI/Wan2.1-Fun-1.3B-InP](https://modelscope.cn/models/PAI/Wan2.1-Fun-1.3B-InP)|`input_image`, `end_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.1-Fun-1.3B-InP.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.1-Fun-1.3B-InP.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.1-Fun-1.3B-InP.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.1-Fun-1.3B-InP.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.1-Fun-1.3B-InP.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.1-Fun-1.3B-InP.py)|
-|[PAI/Wan2.1-Fun-1.3B-Control](https://modelscope.cn/models/PAI/Wan2.1-Fun-1.3B-Control)|`control_video`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.1-Fun-1.3B-Control.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.1-Fun-1.3B-Control.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.1-Fun-1.3B-Control.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.1-Fun-1.3B-Control.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.1-Fun-1.3B-Control.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.1-Fun-1.3B-Control.py)|
-|[PAI/Wan2.1-Fun-14B-InP](https://modelscope.cn/models/PAI/Wan2.1-Fun-14B-InP)|`input_image`, `end_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.1-Fun-14B-InP.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.1-Fun-14B-InP.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.1-Fun-14B-InP.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.1-Fun-14B-InP.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.1-Fun-14B-InP.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.1-Fun-14B-InP.py)|
-|[PAI/Wan2.1-Fun-14B-Control](https://modelscope.cn/models/PAI/Wan2.1-Fun-14B-Control)|`control_video`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.1-Fun-14B-Control.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.1-Fun-14B-Control.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.1-Fun-14B-Control.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.1-Fun-14B-Control.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.1-Fun-14B-Control.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.1-Fun-14B-Control.py)|
-|[PAI/Wan2.1-Fun-V1.1-1.3B-Control](https://modelscope.cn/models/PAI/Wan2.1-Fun-V1.1-1.3B-Control)|`control_video`, `reference_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.1-Fun-V1.1-1.3B-Control.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.1-Fun-V1.1-1.3B-Control.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.1-Fun-V1.1-1.3B-Control.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.1-Fun-V1.1-1.3B-Control.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.1-Fun-V1.1-1.3B-Control.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.1-Fun-V1.1-1.3B-Control.py)|
-|[PAI/Wan2.1-Fun-V1.1-14B-Control](https://modelscope.cn/models/PAI/Wan2.1-Fun-V1.1-14B-Control)|`control_video`, `reference_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.1-Fun-V1.1-14B-Control.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.1-Fun-V1.1-14B-Control.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.1-Fun-V1.1-14B-Control.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.1-Fun-V1.1-14B-Control.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.1-Fun-V1.1-14B-Control.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.1-Fun-V1.1-14B-Control.py)|
-|[PAI/Wan2.1-Fun-V1.1-1.3B-InP](https://modelscope.cn/models/PAI/Wan2.1-Fun-V1.1-1.3B-InP)|`input_image`, `end_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.1-Fun-V1.1-1.3B-InP.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.1-Fun-V1.1-1.3B-InP.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.1-Fun-V1.1-1.3B-InP.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.1-Fun-V1.1-1.3B-InP.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.1-Fun-V1.1-1.3B-InP.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.1-Fun-V1.1-1.3B-InP.py)|
-|[PAI/Wan2.1-Fun-V1.1-14B-InP](https://modelscope.cn/models/PAI/Wan2.1-Fun-V1.1-14B-InP)|`input_image`, `end_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.1-Fun-V1.1-14B-InP.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.1-Fun-V1.1-14B-InP.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.1-Fun-V1.1-14B-InP.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.1-Fun-V1.1-14B-InP.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.1-Fun-V1.1-14B-InP.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.1-Fun-V1.1-14B-InP.py)|
-|[PAI/Wan2.1-Fun-V1.1-1.3B-Control-Camera](https://modelscope.cn/models/PAI/Wan2.1-Fun-V1.1-1.3B-Control-Camera)|`control_camera_video`, `input_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.1-Fun-V1.1-1.3B-Control-Camera.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.1-Fun-V1.1-1.3B-Control-Camera.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.1-Fun-V1.1-1.3B-Control-Camera.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.1-Fun-V1.1-1.3B-Control-Camera.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.1-Fun-V1.1-1.3B-Control-Camera.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.1-Fun-V1.1-1.3B-Control-Camera.py)|
-|[PAI/Wan2.1-Fun-V1.1-14B-Control-Camera](https://modelscope.cn/models/PAI/Wan2.1-Fun-V1.1-14B-Control-Camera)|`control_camera_video`, `input_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.1-Fun-V1.1-14B-Control-Camera.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.1-Fun-V1.1-14B-Control-Camera.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.1-Fun-V1.1-14B-Control-Camera.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.1-Fun-V1.1-14B-Control-Camera.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.1-Fun-V1.1-14B-Control-Camera.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.1-Fun-V1.1-14B-Control-Camera.py)|
-|[DiffSynth-Studio/Wan2.1-1.3b-speedcontrol-v1](https://modelscope.cn/models/DiffSynth-Studio/Wan2.1-1.3b-speedcontrol-v1)|`motion_bucket_id`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.1-1.3b-speedcontrol-v1.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.1-1.3b-speedcontrol-v1.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.1-1.3b-speedcontrol-v1.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.1-1.3b-speedcontrol-v1.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.1-1.3b-speedcontrol-v1.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.1-1.3b-speedcontrol-v1.py)|
-|[krea/krea-realtime-video](https://www.modelscope.cn/models/krea/krea-realtime-video)||[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/krea-realtime-video.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/krea-realtime-video.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/krea-realtime-video.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/krea-realtime-video.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/krea-realtime-video.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/krea-realtime-video.py)|
-|[meituan-longcat/LongCat-Video](https://www.modelscope.cn/models/meituan-longcat/LongCat-Video)|`longcat_video`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/LongCat-Video.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/LongCat-Video.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/LongCat-Video.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/LongCat-Video.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/LongCat-Video.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/LongCat-Video.py)|
-|[ByteDance/Video-As-Prompt-Wan2.1-14B](https://modelscope.cn/models/ByteDance/Video-As-Prompt-Wan2.1-14B)|`vap_video`, `vap_prompt`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Video-As-Prompt-Wan2.1-14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Video-As-Prompt-Wan2.1-14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Video-As-Prompt-Wan2.1-14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Video-As-Prompt-Wan2.1-14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Video-As-Prompt-Wan2.1-14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Video-As-Prompt-Wan2.1-14B.py)|
-|[Wan-AI/Wan2.2-T2V-A14B](https://modelscope.cn/models/Wan-AI/Wan2.2-T2V-A14B)||[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.2-T2V-A14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.2-T2V-A14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.2-T2V-A14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.2-T2V-A14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.2-T2V-A14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.2-T2V-A14B.py)|
-|[Wan-AI/Wan2.2-I2V-A14B](https://modelscope.cn/models/Wan-AI/Wan2.2-I2V-A14B)|`input_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.2-I2V-A14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.2-I2V-A14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.2-I2V-A14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.2-I2V-A14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.2-I2V-A14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.2-I2V-A14B.py)|
-|[Wan-AI/Wan2.2-TI2V-5B](https://modelscope.cn/models/Wan-AI/Wan2.2-TI2V-5B)|`input_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.2-TI2V-5B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.2-TI2V-5B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.2-TI2V-5B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.2-TI2V-5B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.2-TI2V-5B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.2-TI2V-5B.py)|
-|[Wan-AI/Wan2.2-Animate-14B](https://www.modelscope.cn/models/Wan-AI/Wan2.2-Animate-14B)|`input_image`, `animate_pose_video`, `animate_face_video`, `animate_inpaint_video`, `animate_mask_video`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.2-Animate-14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.2-Animate-14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.2-Animate-14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.2-Animate-14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.2-Animate-14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.2-Animate-14B.py)|
-|[Wan-AI/Wan2.2-S2V-14B](https://www.modelscope.cn/models/Wan-AI/Wan2.2-S2V-14B)|`input_image`, `input_audio`, `audio_sample_rate`, `s2v_pose_video`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.2-S2V-14B_multi_clips.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.2-S2V-14B_multi_clips.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.2-S2V-14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.2-S2V-14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.2-S2V-14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.2-S2V-14B.py)|
-|[PAI/Wan2.2-VACE-Fun-A14B](https://www.modelscope.cn/models/PAI/Wan2.2-VACE-Fun-A14B)|`vace_control_video`, `vace_reference_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.2-VACE-Fun-A14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.2-VACE-Fun-A14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.2-VACE-Fun-A14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.2-VACE-Fun-A14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.2-VACE-Fun-A14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.2-VACE-Fun-A14B.py)|
-|[PAI/Wan2.2-Fun-A14B-InP](https://modelscope.cn/models/PAI/Wan2.2-Fun-A14B-InP)|`input_image`, `end_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.2-Fun-A14B-InP.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.2-Fun-A14B-InP.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.2-Fun-A14B-InP.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.2-Fun-A14B-InP.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.2-Fun-A14B-InP.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.2-Fun-A14B-InP.py)|
-|[PAI/Wan2.2-Fun-A14B-Control](https://modelscope.cn/models/PAI/Wan2.2-Fun-A14B-Control)|`control_video`, `reference_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.2-Fun-A14B-Control.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.2-Fun-A14B-Control.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.2-Fun-A14B-Control.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.2-Fun-A14B-Control.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.2-Fun-A14B-Control.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.2-Fun-A14B-Control.py)|
-|[PAI/Wan2.2-Fun-A14B-Control-Camera](https://modelscope.cn/models/PAI/Wan2.2-Fun-A14B-Control-Camera)|`control_camera_video`, `input_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.2-Fun-A14B-Control-Camera.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.2-Fun-A14B-Control-Camera.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.2-Fun-A14B-Control-Camera.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.2-Fun-A14B-Control-Camera.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.2-Fun-A14B-Control-Camera.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.2-Fun-A14B-Control-Camera.py)|
-|[openmoss/MOVA-360p](https://modelscope.cn/models/openmoss/MOVA-360p)|`input_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/mova/model_inference/MOVA-360p-I2AV.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/mova/model_inference_low_vram/MOVA-360p-I2AV.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/mova/model_training/full/MOVA-360P-I2AV.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/mova/model_training/validate_full/MOVA-360p-I2AV.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/mova/model_training/lora/MOVA-360P-I2AV.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/mova/model_training/validate_lora/MOVA-360p-I2AV.py)|
-|[openmoss/MOVA-720p](https://modelscope.cn/models/openmoss/MOVA-720p)|`input_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/mova/model_inference/MOVA-720p-I2AV.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/mova/model_inference_low_vram/MOVA-720p-I2AV.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/mova/model_training/full/MOVA-720P-I2AV.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/mova/model_training/validate_full/MOVA-720p-I2AV.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/mova/model_training/lora/MOVA-720P-I2AV.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/mova/model_training/validate_lora/MOVA-720p-I2AV.py)|
-|[Wan-AI/Wan-Dancer-14B (global model)](https://modelscope.cn/models/Wan-AI/Wan-Dancer-14B)|`wantodance_music_path`, `wantodance_reference_image`, `wantodance_fps`, `wantodance_keyframes`, `wantodance_keyframes_mask`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan-Dancer-14B-global.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan-Dancer-14B-global.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan-Dancer-14B-global.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan-Dancer-14B-global.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan-Dancer-14B-global.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan-Dancer-14B-global.py)|
-|[Wan-AI/Wan-Dancer-14B (local model)](https://modelscope.cn/models/Wan-AI/Wan-Dancer-14B)|`wantodance_music_path`, `wantodance_reference_image`, `wantodance_fps`, `wantodance_keyframes`, `wantodance_keyframes_mask`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan-Dancer-14B-local.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan-Dancer-14B-local.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan-Dancer-14B-local.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan-Dancer-14B-local.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan-Dancer-14B-local.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan-Dancer-14B-local.py)|
+|[Wan-AI/Wan2.1-T2V-1.3B](https://modelscope.cn/models/Wan-AI/Wan2.1-T2V-1.3B)||[code](/examples/wanvideo/model_inference/Wan2.1-T2V-1.3B.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.1-T2V-1.3B.py)|[code](/examples/wanvideo/model_training/full/Wan2.1-T2V-1.3B.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.1-T2V-1.3B.py)|[code](/examples/wanvideo/model_training/lora/Wan2.1-T2V-1.3B.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.1-T2V-1.3B.py)|
+|[Wan-AI/Wan2.1-T2V-14B](https://modelscope.cn/models/Wan-AI/Wan2.1-T2V-14B)||[code](/examples/wanvideo/model_inference/Wan2.1-T2V-14B.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.1-T2V-14B.py)|[code](/examples/wanvideo/model_training/full/Wan2.1-T2V-14B.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.1-T2V-14B.py)|[code](/examples/wanvideo/model_training/lora/Wan2.1-T2V-14B.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.1-T2V-14B.py)|
+|[Wan-AI/Wan2.1-I2V-14B-480P](https://modelscope.cn/models/Wan-AI/Wan2.1-I2V-14B-480P)|`input_image`|[code](/examples/wanvideo/model_inference/Wan2.1-I2V-14B-480P.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.1-I2V-14B-480P.py)|[code](/examples/wanvideo/model_training/full/Wan2.1-I2V-14B-480P.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.1-I2V-14B-480P.py)|[code](/examples/wanvideo/model_training/lora/Wan2.1-I2V-14B-480P.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.1-I2V-14B-480P.py)|
+|[Wan-AI/Wan2.1-I2V-14B-720P](https://modelscope.cn/models/Wan-AI/Wan2.1-I2V-14B-720P)|`input_image`|[code](/examples/wanvideo/model_inference/Wan2.1-I2V-14B-720P.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.1-I2V-14B-720P.py)|[code](/examples/wanvideo/model_training/full/Wan2.1-I2V-14B-720P.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.1-I2V-14B-720P.py)|[code](/examples/wanvideo/model_training/lora/Wan2.1-I2V-14B-720P.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.1-I2V-14B-720P.py)|
+|[Wan-AI/Wan2.1-FLF2V-14B-720P](https://modelscope.cn/models/Wan-AI/Wan2.1-FLF2V-14B-720P)|`input_image`, `end_image`|[code](/examples/wanvideo/model_inference/Wan2.1-FLF2V-14B-720P.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.1-FLF2V-14B-720P.py)|[code](/examples/wanvideo/model_training/full/Wan2.1-FLF2V-14B-720P.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.1-FLF2V-14B-720P.py)|[code](/examples/wanvideo/model_training/lora/Wan2.1-FLF2V-14B-720P.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.1-FLF2V-14B-720P.py)|
+|[iic/VACE-Wan2.1-1.3B-Preview](https://modelscope.cn/models/iic/VACE-Wan2.1-1.3B-Preview)|`vace_control_video`, `vace_reference_image`|[code](/examples/wanvideo/model_inference/Wan2.1-VACE-1.3B-Preview.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.1-VACE-1.3B-Preview.py)|[code](/examples/wanvideo/model_training/full/Wan2.1-VACE-1.3B-Preview.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.1-VACE-1.3B-Preview.py)|[code](/examples/wanvideo/model_training/lora/Wan2.1-VACE-1.3B-Preview.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.1-VACE-1.3B-Preview.py)|
+|[Wan-AI/Wan2.1-VACE-1.3B](https://modelscope.cn/models/Wan-AI/Wan2.1-VACE-1.3B)|`vace_control_video`, `vace_reference_image`|[code](/examples/wanvideo/model_inference/Wan2.1-VACE-1.3B.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.1-VACE-1.3B.py)|[code](/examples/wanvideo/model_training/full/Wan2.1-VACE-1.3B.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.1-VACE-1.3B.py)|[code](/examples/wanvideo/model_training/lora/Wan2.1-VACE-1.3B.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.1-VACE-1.3B.py)|
+|[Wan-AI/Wan2.1-VACE-14B](https://modelscope.cn/models/Wan-AI/Wan2.1-VACE-14B)|`vace_control_video`, `vace_reference_image`|[code](/examples/wanvideo/model_inference/Wan2.1-VACE-14B.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.1-VACE-14B.py)|[code](/examples/wanvideo/model_training/full/Wan2.1-VACE-14B.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.1-VACE-14B.py)|[code](/examples/wanvideo/model_training/lora/Wan2.1-VACE-14B.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.1-VACE-14B.py)|
+|[PAI/Wan2.1-Fun-1.3B-InP](https://modelscope.cn/models/PAI/Wan2.1-Fun-1.3B-InP)|`input_image`, `end_image`|[code](/examples/wanvideo/model_inference/Wan2.1-Fun-1.3B-InP.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.1-Fun-1.3B-InP.py)|[code](/examples/wanvideo/model_training/full/Wan2.1-Fun-1.3B-InP.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.1-Fun-1.3B-InP.py)|[code](/examples/wanvideo/model_training/lora/Wan2.1-Fun-1.3B-InP.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.1-Fun-1.3B-InP.py)|
+|[PAI/Wan2.1-Fun-1.3B-Control](https://modelscope.cn/models/PAI/Wan2.1-Fun-1.3B-Control)|`control_video`|[code](/examples/wanvideo/model_inference/Wan2.1-Fun-1.3B-Control.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.1-Fun-1.3B-Control.py)|[code](/examples/wanvideo/model_training/full/Wan2.1-Fun-1.3B-Control.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.1-Fun-1.3B-Control.py)|[code](/examples/wanvideo/model_training/lora/Wan2.1-Fun-1.3B-Control.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.1-Fun-1.3B-Control.py)|
+|[PAI/Wan2.1-Fun-14B-InP](https://modelscope.cn/models/PAI/Wan2.1-Fun-14B-InP)|`input_image`, `end_image`|[code](/examples/wanvideo/model_inference/Wan2.1-Fun-14B-InP.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.1-Fun-14B-InP.py)|[code](/examples/wanvideo/model_training/full/Wan2.1-Fun-14B-InP.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.1-Fun-14B-InP.py)|[code](/examples/wanvideo/model_training/lora/Wan2.1-Fun-14B-InP.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.1-Fun-14B-InP.py)|
+|[PAI/Wan2.1-Fun-14B-Control](https://modelscope.cn/models/PAI/Wan2.1-Fun-14B-Control)|`control_video`|[code](/examples/wanvideo/model_inference/Wan2.1-Fun-14B-Control.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.1-Fun-14B-Control.py)|[code](/examples/wanvideo/model_training/full/Wan2.1-Fun-14B-Control.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.1-Fun-14B-Control.py)|[code](/examples/wanvideo/model_training/lora/Wan2.1-Fun-14B-Control.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.1-Fun-14B-Control.py)|
+|[PAI/Wan2.1-Fun-V1.1-1.3B-Control](https://modelscope.cn/models/PAI/Wan2.1-Fun-V1.1-1.3B-Control)|`control_video`, `reference_image`|[code](/examples/wanvideo/model_inference/Wan2.1-Fun-V1.1-1.3B-Control.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.1-Fun-V1.1-1.3B-Control.py)|[code](/examples/wanvideo/model_training/full/Wan2.1-Fun-V1.1-1.3B-Control.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.1-Fun-V1.1-1.3B-Control.py)|[code](/examples/wanvideo/model_training/lora/Wan2.1-Fun-V1.1-1.3B-Control.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.1-Fun-V1.1-1.3B-Control.py)|
+|[PAI/Wan2.1-Fun-V1.1-14B-Control](https://modelscope.cn/models/PAI/Wan2.1-Fun-V1.1-14B-Control)|`control_video`, `reference_image`|[code](/examples/wanvideo/model_inference/Wan2.1-Fun-V1.1-14B-Control.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.1-Fun-V1.1-14B-Control.py)|[code](/examples/wanvideo/model_training/full/Wan2.1-Fun-V1.1-14B-Control.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.1-Fun-V1.1-14B-Control.py)|[code](/examples/wanvideo/model_training/lora/Wan2.1-Fun-V1.1-14B-Control.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.1-Fun-V1.1-14B-Control.py)|
+|[PAI/Wan2.1-Fun-V1.1-1.3B-InP](https://modelscope.cn/models/PAI/Wan2.1-Fun-V1.1-1.3B-InP)|`input_image`, `end_image`|[code](/examples/wanvideo/model_inference/Wan2.1-Fun-V1.1-1.3B-InP.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.1-Fun-V1.1-1.3B-InP.py)|[code](/examples/wanvideo/model_training/full/Wan2.1-Fun-V1.1-1.3B-InP.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.1-Fun-V1.1-1.3B-InP.py)|[code](/examples/wanvideo/model_training/lora/Wan2.1-Fun-V1.1-1.3B-InP.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.1-Fun-V1.1-1.3B-InP.py)|
+|[PAI/Wan2.1-Fun-V1.1-14B-InP](https://modelscope.cn/models/PAI/Wan2.1-Fun-V1.1-14B-InP)|`input_image`, `end_image`|[code](/examples/wanvideo/model_inference/Wan2.1-Fun-V1.1-14B-InP.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.1-Fun-V1.1-14B-InP.py)|[code](/examples/wanvideo/model_training/full/Wan2.1-Fun-V1.1-14B-InP.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.1-Fun-V1.1-14B-InP.py)|[code](/examples/wanvideo/model_training/lora/Wan2.1-Fun-V1.1-14B-InP.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.1-Fun-V1.1-14B-InP.py)|
+|[PAI/Wan2.1-Fun-V1.1-1.3B-Control-Camera](https://modelscope.cn/models/PAI/Wan2.1-Fun-V1.1-1.3B-Control-Camera)|`control_camera_video`, `input_image`|[code](/examples/wanvideo/model_inference/Wan2.1-Fun-V1.1-1.3B-Control-Camera.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.1-Fun-V1.1-1.3B-Control-Camera.py)|[code](/examples/wanvideo/model_training/full/Wan2.1-Fun-V1.1-1.3B-Control-Camera.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.1-Fun-V1.1-1.3B-Control-Camera.py)|[code](/examples/wanvideo/model_training/lora/Wan2.1-Fun-V1.1-1.3B-Control-Camera.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.1-Fun-V1.1-1.3B-Control-Camera.py)|
+|[PAI/Wan2.1-Fun-V1.1-14B-Control-Camera](https://modelscope.cn/models/PAI/Wan2.1-Fun-V1.1-14B-Control-Camera)|`control_camera_video`, `input_image`|[code](/examples/wanvideo/model_inference/Wan2.1-Fun-V1.1-14B-Control-Camera.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.1-Fun-V1.1-14B-Control-Camera.py)|[code](/examples/wanvideo/model_training/full/Wan2.1-Fun-V1.1-14B-Control-Camera.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.1-Fun-V1.1-14B-Control-Camera.py)|[code](/examples/wanvideo/model_training/lora/Wan2.1-Fun-V1.1-14B-Control-Camera.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.1-Fun-V1.1-14B-Control-Camera.py)|
+|[DiffSynth-Studio/Wan2.1-1.3b-speedcontrol-v1](https://modelscope.cn/models/DiffSynth-Studio/Wan2.1-1.3b-speedcontrol-v1)|`motion_bucket_id`|[code](/examples/wanvideo/model_inference/Wan2.1-1.3b-speedcontrol-v1.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.1-1.3b-speedcontrol-v1.py)|[code](/examples/wanvideo/model_training/full/Wan2.1-1.3b-speedcontrol-v1.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.1-1.3b-speedcontrol-v1.py)|[code](/examples/wanvideo/model_training/lora/Wan2.1-1.3b-speedcontrol-v1.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.1-1.3b-speedcontrol-v1.py)|
+|[krea/krea-realtime-video](https://www.modelscope.cn/models/krea/krea-realtime-video)||[code](/examples/wanvideo/model_inference/krea-realtime-video.py)|[code](/examples/wanvideo/model_inference_low_vram/krea-realtime-video.py)|[code](/examples/wanvideo/model_training/full/krea-realtime-video.sh)|[code](/examples/wanvideo/model_training/validate_full/krea-realtime-video.py)|[code](/examples/wanvideo/model_training/lora/krea-realtime-video.sh)|[code](/examples/wanvideo/model_training/validate_lora/krea-realtime-video.py)|
+|[meituan-longcat/LongCat-Video](https://www.modelscope.cn/models/meituan-longcat/LongCat-Video)|`longcat_video`|[code](/examples/wanvideo/model_inference/LongCat-Video.py)|[code](/examples/wanvideo/model_inference_low_vram/LongCat-Video.py)|[code](/examples/wanvideo/model_training/full/LongCat-Video.sh)|[code](/examples/wanvideo/model_training/validate_full/LongCat-Video.py)|[code](/examples/wanvideo/model_training/lora/LongCat-Video.sh)|[code](/examples/wanvideo/model_training/validate_lora/LongCat-Video.py)|
+|[ByteDance/Video-As-Prompt-Wan2.1-14B](https://modelscope.cn/models/ByteDance/Video-As-Prompt-Wan2.1-14B)|`vap_video`, `vap_prompt`|[code](/examples/wanvideo/model_inference/Video-As-Prompt-Wan2.1-14B.py)|[code](/examples/wanvideo/model_inference_low_vram/Video-As-Prompt-Wan2.1-14B.py)|[code](/examples/wanvideo/model_training/full/Video-As-Prompt-Wan2.1-14B.sh)|[code](/examples/wanvideo/model_training/validate_full/Video-As-Prompt-Wan2.1-14B.py)|[code](/examples/wanvideo/model_training/lora/Video-As-Prompt-Wan2.1-14B.sh)|[code](/examples/wanvideo/model_training/validate_lora/Video-As-Prompt-Wan2.1-14B.py)|
+|[Wan-AI/Wan2.2-T2V-A14B](https://modelscope.cn/models/Wan-AI/Wan2.2-T2V-A14B)||[code](/examples/wanvideo/model_inference/Wan2.2-T2V-A14B.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.2-T2V-A14B.py)|[code](/examples/wanvideo/model_training/full/Wan2.2-T2V-A14B.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.2-T2V-A14B.py)|[code](/examples/wanvideo/model_training/lora/Wan2.2-T2V-A14B.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.2-T2V-A14B.py)|
+|[Wan-AI/Wan2.2-I2V-A14B](https://modelscope.cn/models/Wan-AI/Wan2.2-I2V-A14B)|`input_image`|[code](/examples/wanvideo/model_inference/Wan2.2-I2V-A14B.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.2-I2V-A14B.py)|[code](/examples/wanvideo/model_training/full/Wan2.2-I2V-A14B.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.2-I2V-A14B.py)|[code](/examples/wanvideo/model_training/lora/Wan2.2-I2V-A14B.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.2-I2V-A14B.py)|
+|[Wan-AI/Wan2.2-TI2V-5B](https://modelscope.cn/models/Wan-AI/Wan2.2-TI2V-5B)|`input_image`|[code](/examples/wanvideo/model_inference/Wan2.2-TI2V-5B.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.2-TI2V-5B.py)|[code](/examples/wanvideo/model_training/full/Wan2.2-TI2V-5B.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.2-TI2V-5B.py)|[code](/examples/wanvideo/model_training/lora/Wan2.2-TI2V-5B.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.2-TI2V-5B.py)|
+|[Wan-AI/Wan2.2-Animate-14B](https://www.modelscope.cn/models/Wan-AI/Wan2.2-Animate-14B)|`input_image`, `animate_pose_video`, `animate_face_video`, `animate_inpaint_video`, `animate_mask_video`|[code](/examples/wanvideo/model_inference/Wan2.2-Animate-14B.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.2-Animate-14B.py)|[code](/examples/wanvideo/model_training/full/Wan2.2-Animate-14B.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.2-Animate-14B.py)|[code](/examples/wanvideo/model_training/lora/Wan2.2-Animate-14B.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.2-Animate-14B.py)|
+|[Wan-AI/Wan2.2-Animate-2-14B](https://www.modelscope.cn/models/Wan-AI/Wan2.2-Animate-2-14B)|`animate2_reference_image`, `animate2_reference_video`, `animate2_prompt_ref`|[code](/examples/wanvideo/model_inference/Wan2.2-Animate-2-14B.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.2-Animate-2-14B.py)|[code](/examples/wanvideo/model_training/full/Wan2.2-Animate-2-14B.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.2-Animate-2-14B.py)|[code](/examples/wanvideo/model_training/lora/Wan2.2-Animate-2-14B.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.2-Animate-2-14B.py)|
+|[Wan-AI/Wan2.2-Animate-2-14B: Distilled](https://www.modelscope.cn/models/Wan-AI/Wan2.2-Animate-2-14B)|`animate2_reference_image`, `animate2_reference_video`, `animate2_prompt_ref`|[code](/examples/wanvideo/model_inference/Wan2.2-Animate-2-14B-Distilled.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.2-Animate-2-14B-Distilled.py)|[code](/examples/wanvideo/model_training/full/Wan2.2-Animate-2-14B-Distilled.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.2-Animate-2-14B-Distilled.py)|[code](/examples/wanvideo/model_training/lora/Wan2.2-Animate-2-14B-Distilled.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.2-Animate-2-14B-Distilled.py)|
+|[Wan-AI/Wan2.2-S2V-14B](https://www.modelscope.cn/models/Wan-AI/Wan2.2-S2V-14B)|`input_image`, `input_audio`, `audio_sample_rate`, `s2v_pose_video`|[code](/examples/wanvideo/model_inference/Wan2.2-S2V-14B_multi_clips.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.2-S2V-14B_multi_clips.py)|[code](/examples/wanvideo/model_training/full/Wan2.2-S2V-14B.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.2-S2V-14B.py)|[code](/examples/wanvideo/model_training/lora/Wan2.2-S2V-14B.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.2-S2V-14B.py)|
+|[PAI/Wan2.2-VACE-Fun-A14B](https://www.modelscope.cn/models/PAI/Wan2.2-VACE-Fun-A14B)|`vace_control_video`, `vace_reference_image`|[code](/examples/wanvideo/model_inference/Wan2.2-VACE-Fun-A14B.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.2-VACE-Fun-A14B.py)|[code](/examples/wanvideo/model_training/full/Wan2.2-VACE-Fun-A14B.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.2-VACE-Fun-A14B.py)|[code](/examples/wanvideo/model_training/lora/Wan2.2-VACE-Fun-A14B.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.2-VACE-Fun-A14B.py)|
+|[PAI/Wan2.2-Fun-A14B-InP](https://modelscope.cn/models/PAI/Wan2.2-Fun-A14B-InP)|`input_image`, `end_image`|[code](/examples/wanvideo/model_inference/Wan2.2-Fun-A14B-InP.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.2-Fun-A14B-InP.py)|[code](/examples/wanvideo/model_training/full/Wan2.2-Fun-A14B-InP.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.2-Fun-A14B-InP.py)|[code](/examples/wanvideo/model_training/lora/Wan2.2-Fun-A14B-InP.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.2-Fun-A14B-InP.py)|
+|[PAI/Wan2.2-Fun-A14B-Control](https://modelscope.cn/models/PAI/Wan2.2-Fun-A14B-Control)|`control_video`, `reference_image`|[code](/examples/wanvideo/model_inference/Wan2.2-Fun-A14B-Control.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.2-Fun-A14B-Control.py)|[code](/examples/wanvideo/model_training/full/Wan2.2-Fun-A14B-Control.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.2-Fun-A14B-Control.py)|[code](/examples/wanvideo/model_training/lora/Wan2.2-Fun-A14B-Control.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.2-Fun-A14B-Control.py)|
+|[PAI/Wan2.2-Fun-A14B-Control-Camera](https://modelscope.cn/models/PAI/Wan2.2-Fun-A14B-Control-Camera)|`control_camera_video`, `input_image`|[code](/examples/wanvideo/model_inference/Wan2.2-Fun-A14B-Control-Camera.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.2-Fun-A14B-Control-Camera.py)|[code](/examples/wanvideo/model_training/full/Wan2.2-Fun-A14B-Control-Camera.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.2-Fun-A14B-Control-Camera.py)|[code](/examples/wanvideo/model_training/lora/Wan2.2-Fun-A14B-Control-Camera.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.2-Fun-A14B-Control-Camera.py)|
+|[openmoss/MOVA-360p](https://modelscope.cn/models/openmoss/MOVA-360p)|`input_image`|[code](/examples/mova/model_inference/MOVA-360p-I2AV.py)|[code](/examples/mova/model_inference_low_vram/MOVA-360p-I2AV.py)|[code](/examples/mova/model_training/full/MOVA-360P-I2AV.sh)|[code](/examples/mova/model_training/validate_full/MOVA-360p-I2AV.py)|[code](/examples/mova/model_training/lora/MOVA-360P-I2AV.sh)|[code](/examples/mova/model_training/validate_lora/MOVA-360p-I2AV.py)|
+|[openmoss/MOVA-720p](https://modelscope.cn/models/openmoss/MOVA-720p)|`input_image`|[code](/examples/mova/model_inference/MOVA-720p-I2AV.py)|[code](/examples/mova/model_inference_low_vram/MOVA-720p-I2AV.py)|[code](/examples/mova/model_training/full/MOVA-720P-I2AV.sh)|[code](/examples/mova/model_training/validate_full/MOVA-720p-I2AV.py)|[code](/examples/mova/model_training/lora/MOVA-720P-I2AV.sh)|[code](/examples/mova/model_training/validate_lora/MOVA-720p-I2AV.py)|
+|[Wan-AI/Wan-Dancer-14B (global model)](https://modelscope.cn/models/Wan-AI/Wan-Dancer-14B)|`wantodance_music_path`, `wantodance_reference_image`, `wantodance_fps`, `wantodance_keyframes`, `wantodance_keyframes_mask`|[code](/examples/wanvideo/model_inference/Wan-Dancer-14B-global.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan-Dancer-14B-global.py)|[code](/examples/wanvideo/model_training/full/Wan-Dancer-14B-global.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan-Dancer-14B-global.py)|[code](/examples/wanvideo/model_training/lora/Wan-Dancer-14B-global.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan-Dancer-14B-global.py)|
+|[Wan-AI/Wan-Dancer-14B (local model)](https://modelscope.cn/models/Wan-AI/Wan-Dancer-14B)|`wantodance_music_path`, `wantodance_reference_image`, `wantodance_fps`, `wantodance_keyframes`, `wantodance_keyframes_mask`|[code](/examples/wanvideo/model_inference/Wan-Dancer-14B-local.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan-Dancer-14B-local.py)|[code](/examples/wanvideo/model_training/full/Wan-Dancer-14B-local.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan-Dancer-14B-local.py)|[code](/examples/wanvideo/model_training/lora/Wan-Dancer-14B-local.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan-Dancer-14B-local.py)|
 
 </details>
 
diff --git a/README_zh.md b/README_zh.md
index 91f7e6e90..58ee31fb9 100644
--- a/README_zh.md
+++ b/README_zh.md
@@ -36,6 +36,8 @@ DiffSynth 目前包括两个开源项目：
 
 > 目前本项目的开发人员有限，大部分工作由 [Artiprocher](https://github.com/Artiprocher) 和 [mi804](https://github.com/mi804) 负责，因此新功能的开发进展会比较缓慢，issue 的回复和解决速度有限，我们对此感到非常抱歉，请各位开发者理解。
 
+- **2026年8月7日** 我们为 Wan 系列新增了 Wan-Animate-2，输入一张参考图和一段驱动视频，即可让参考角色演绎驱动视频中的动作，生成高质量角色动画，包含标准与蒸馏两个变体。详情请参考[文档](/docs/zh/Model_Details/Wan.md)和[示例代码](/examples/wanvideo/)。
+
 - **2026年8月3日** MiniMax-H3 开源，欢迎加入视频生成模型家族！支持文生音视频、首尾帧引导生成、参考驱动生成、低显存推理以及 NF4 量化推理。详情请参考[文档](/docs/zh/Model_Details/MiniMax-H3.md)和[示例代码](/examples/minimax_h3/)。
 
 - **2026年7月28日** LingBot-Video 开源，欢迎加入视频生成模型家族！本次开源包含 Dense-1.3B 和 MoE-30B-A3B 两个版本（MoE 版总参数量 30B、每个 token 激活约 3B），均支持文生视频、图生视频、文生图推理、低显存推理以及 LoRA / 全量训练能力。详情请参考[文档](/docs/zh/Model_Details/LingBot-Video.md)和[示例代码](/examples/lingbot_video/)。特别感谢 [NancyFyong](https://github.com/NancyFyong) 贡献了本模型的接入代码！
@@ -1413,41 +1415,43 @@ Wan 的示例代码位于：[/examples/wanvideo/](/examples/wanvideo/)
 
 |模型 ID|额外参数|推理|低显存推理|全量训练|全量训练后验证|LoRA 训练|LoRA 训练后验证|
 |-|-|-|-|-|-|-|-|
-|[Wan-AI/Wan2.1-T2V-1.3B](https://modelscope.cn/models/Wan-AI/Wan2.1-T2V-1.3B)||[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.1-T2V-1.3B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.1-T2V-1.3B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.1-T2V-1.3B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.1-T2V-1.3B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.1-T2V-1.3B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.1-T2V-1.3B.py)|
-|[Wan-AI/Wan2.1-T2V-14B](https://modelscope.cn/models/Wan-AI/Wan2.1-T2V-14B)||[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.1-T2V-14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.1-T2V-14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.1-T2V-14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.1-T2V-14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.1-T2V-14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.1-T2V-14B.py)|
-|[Wan-AI/Wan2.1-I2V-14B-480P](https://modelscope.cn/models/Wan-AI/Wan2.1-I2V-14B-480P)|`input_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.1-I2V-14B-480P.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.1-I2V-14B-480P.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.1-I2V-14B-480P.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.1-I2V-14B-480P.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.1-I2V-14B-480P.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.1-I2V-14B-480P.py)|
-|[Wan-AI/Wan2.1-I2V-14B-720P](https://modelscope.cn/models/Wan-AI/Wan2.1-I2V-14B-720P)|`input_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.1-I2V-14B-720P.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.1-I2V-14B-720P.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.1-I2V-14B-720P.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.1-I2V-14B-720P.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.1-I2V-14B-720P.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.1-I2V-14B-720P.py)|
-|[Wan-AI/Wan2.1-FLF2V-14B-720P](https://modelscope.cn/models/Wan-AI/Wan2.1-FLF2V-14B-720P)|`input_image`, `end_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.1-FLF2V-14B-720P.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.1-FLF2V-14B-720P.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.1-FLF2V-14B-720P.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.1-FLF2V-14B-720P.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.1-FLF2V-14B-720P.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.1-FLF2V-14B-720P.py)|
-|[iic/VACE-Wan2.1-1.3B-Preview](https://modelscope.cn/models/iic/VACE-Wan2.1-1.3B-Preview)|`vace_control_video`, `vace_reference_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.1-VACE-1.3B-Preview.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.1-VACE-1.3B-Preview.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.1-VACE-1.3B-Preview.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.1-VACE-1.3B-Preview.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.1-VACE-1.3B-Preview.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.1-VACE-1.3B-Preview.py)|
-|[Wan-AI/Wan2.1-VACE-1.3B](https://modelscope.cn/models/Wan-AI/Wan2.1-VACE-1.3B)|`vace_control_video`, `vace_reference_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.1-VACE-1.3B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.1-VACE-1.3B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.1-VACE-1.3B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.1-VACE-1.3B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.1-VACE-1.3B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.1-VACE-1.3B.py)|
-|[Wan-AI/Wan2.1-VACE-14B](https://modelscope.cn/models/Wan-AI/Wan2.1-VACE-14B)|`vace_control_video`, `vace_reference_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.1-VACE-14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.1-VACE-14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.1-VACE-14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.1-VACE-14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.1-VACE-14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.1-VACE-14B.py)|
-|[PAI/Wan2.1-Fun-1.3B-InP](https://modelscope.cn/models/PAI/Wan2.1-Fun-1.3B-InP)|`input_image`, `end_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.1-Fun-1.3B-InP.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.1-Fun-1.3B-InP.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.1-Fun-1.3B-InP.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.1-Fun-1.3B-InP.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.1-Fun-1.3B-InP.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.1-Fun-1.3B-InP.py)|
-|[PAI/Wan2.1-Fun-1.3B-Control](https://modelscope.cn/models/PAI/Wan2.1-Fun-1.3B-Control)|`control_video`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.1-Fun-1.3B-Control.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.1-Fun-1.3B-Control.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.1-Fun-1.3B-Control.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.1-Fun-1.3B-Control.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.1-Fun-1.3B-Control.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.1-Fun-1.3B-Control.py)|
-|[PAI/Wan2.1-Fun-14B-InP](https://modelscope.cn/models/PAI/Wan2.1-Fun-14B-InP)|`input_image`, `end_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.1-Fun-14B-InP.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.1-Fun-14B-InP.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.1-Fun-14B-InP.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.1-Fun-14B-InP.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.1-Fun-14B-InP.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.1-Fun-14B-InP.py)|
-|[PAI/Wan2.1-Fun-14B-Control](https://modelscope.cn/models/PAI/Wan2.1-Fun-14B-Control)|`control_video`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.1-Fun-14B-Control.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.1-Fun-14B-Control.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.1-Fun-14B-Control.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.1-Fun-14B-Control.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.1-Fun-14B-Control.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.1-Fun-14B-Control.py)|
-|[PAI/Wan2.1-Fun-V1.1-1.3B-Control](https://modelscope.cn/models/PAI/Wan2.1-Fun-V1.1-1.3B-Control)|`control_video`, `reference_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.1-Fun-V1.1-1.3B-Control.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.1-Fun-V1.1-1.3B-Control.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.1-Fun-V1.1-1.3B-Control.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.1-Fun-V1.1-1.3B-Control.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.1-Fun-V1.1-1.3B-Control.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.1-Fun-V1.1-1.3B-Control.py)|
-|[PAI/Wan2.1-Fun-V1.1-14B-Control](https://modelscope.cn/models/PAI/Wan2.1-Fun-V1.1-14B-Control)|`control_video`, `reference_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.1-Fun-V1.1-14B-Control.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.1-Fun-V1.1-14B-Control.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.1-Fun-V1.1-14B-Control.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.1-Fun-V1.1-14B-Control.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.1-Fun-V1.1-14B-Control.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.1-Fun-V1.1-14B-Control.py)|
-|[PAI/Wan2.1-Fun-V1.1-1.3B-InP](https://modelscope.cn/models/PAI/Wan2.1-Fun-V1.1-1.3B-InP)|`input_image`, `end_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.1-Fun-V1.1-1.3B-InP.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.1-Fun-V1.1-1.3B-InP.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.1-Fun-V1.1-1.3B-InP.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.1-Fun-V1.1-1.3B-InP.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.1-Fun-V1.1-1.3B-InP.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.1-Fun-V1.1-1.3B-InP.py)|
-|[PAI/Wan2.1-Fun-V1.1-14B-InP](https://modelscope.cn/models/PAI/Wan2.1-Fun-V1.1-14B-InP)|`input_image`, `end_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.1-Fun-V1.1-14B-InP.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.1-Fun-V1.1-14B-InP.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.1-Fun-V1.1-14B-InP.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.1-Fun-V1.1-14B-InP.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.1-Fun-V1.1-14B-InP.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.1-Fun-V1.1-14B-InP.py)|
-|[PAI/Wan2.1-Fun-V1.1-1.3B-Control-Camera](https://modelscope.cn/models/PAI/Wan2.1-Fun-V1.1-1.3B-Control-Camera)|`control_camera_video`, `input_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.1-Fun-V1.1-1.3B-Control-Camera.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.1-Fun-V1.1-1.3B-Control-Camera.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.1-Fun-V1.1-1.3B-Control-Camera.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.1-Fun-V1.1-1.3B-Control-Camera.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.1-Fun-V1.1-1.3B-Control-Camera.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.1-Fun-V1.1-1.3B-Control-Camera.py)|
-|[PAI/Wan2.1-Fun-V1.1-14B-Control-Camera](https://modelscope.cn/models/PAI/Wan2.1-Fun-V1.1-14B-Control-Camera)|`control_camera_video`, `input_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.1-Fun-V1.1-14B-Control-Camera.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.1-Fun-V1.1-14B-Control-Camera.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.1-Fun-V1.1-14B-Control-Camera.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.1-Fun-V1.1-14B-Control-Camera.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.1-Fun-V1.1-14B-Control-Camera.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.1-Fun-V1.1-14B-Control-Camera.py)|
-|[DiffSynth-Studio/Wan2.1-1.3b-speedcontrol-v1](https://modelscope.cn/models/DiffSynth-Studio/Wan2.1-1.3b-speedcontrol-v1)|`motion_bucket_id`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.1-1.3b-speedcontrol-v1.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.1-1.3b-speedcontrol-v1.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.1-1.3b-speedcontrol-v1.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.1-1.3b-speedcontrol-v1.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.1-1.3b-speedcontrol-v1.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.1-1.3b-speedcontrol-v1.py)|
-|[krea/krea-realtime-video](https://www.modelscope.cn/models/krea/krea-realtime-video)||[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/krea-realtime-video.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/krea-realtime-video.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/krea-realtime-video.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/krea-realtime-video.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/krea-realtime-video.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/krea-realtime-video.py)|
-|[meituan-longcat/LongCat-Video](https://www.modelscope.cn/models/meituan-longcat/LongCat-Video)|`longcat_video`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/LongCat-Video.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/LongCat-Video.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/LongCat-Video.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/LongCat-Video.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/LongCat-Video.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/LongCat-Video.py)|
-|[ByteDance/Video-As-Prompt-Wan2.1-14B](https://modelscope.cn/models/ByteDance/Video-As-Prompt-Wan2.1-14B)|`vap_video`, `vap_prompt`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Video-As-Prompt-Wan2.1-14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Video-As-Prompt-Wan2.1-14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Video-As-Prompt-Wan2.1-14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Video-As-Prompt-Wan2.1-14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Video-As-Prompt-Wan2.1-14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Video-As-Prompt-Wan2.1-14B.py)|
-|[Wan-AI/Wan2.2-T2V-A14B](https://modelscope.cn/models/Wan-AI/Wan2.2-T2V-A14B)||[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.2-T2V-A14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.2-T2V-A14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.2-T2V-A14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.2-T2V-A14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.2-T2V-A14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.2-T2V-A14B.py)|
-|[Wan-AI/Wan2.2-I2V-A14B](https://modelscope.cn/models/Wan-AI/Wan2.2-I2V-A14B)|`input_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.2-I2V-A14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.2-I2V-A14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.2-I2V-A14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.2-I2V-A14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.2-I2V-A14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.2-I2V-A14B.py)|
-|[Wan-AI/Wan2.2-TI2V-5B](https://modelscope.cn/models/Wan-AI/Wan2.2-TI2V-5B)|`input_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.2-TI2V-5B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.2-TI2V-5B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.2-TI2V-5B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.2-TI2V-5B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.2-TI2V-5B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.2-TI2V-5B.py)|
-|[Wan-AI/Wan2.2-Animate-14B](https://www.modelscope.cn/models/Wan-AI/Wan2.2-Animate-14B)|`input_image`, `animate_pose_video`, `animate_face_video`, `animate_inpaint_video`, `animate_mask_video`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.2-Animate-14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.2-Animate-14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.2-Animate-14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.2-Animate-14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.2-Animate-14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.2-Animate-14B.py)|
-|[Wan-AI/Wan2.2-S2V-14B](https://www.modelscope.cn/models/Wan-AI/Wan2.2-S2V-14B)|`input_image`, `input_audio`, `audio_sample_rate`, `s2v_pose_video`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.2-S2V-14B_multi_clips.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.2-S2V-14B_multi_clips.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.2-S2V-14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.2-S2V-14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.2-S2V-14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.2-S2V-14B.py)|
-|[PAI/Wan2.2-VACE-Fun-A14B](https://www.modelscope.cn/models/PAI/Wan2.2-VACE-Fun-A14B)|`vace_control_video`, `vace_reference_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.2-VACE-Fun-A14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.2-VACE-Fun-A14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.2-VACE-Fun-A14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.2-VACE-Fun-A14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.2-VACE-Fun-A14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.2-VACE-Fun-A14B.py)|
-|[PAI/Wan2.2-Fun-A14B-InP](https://modelscope.cn/models/PAI/Wan2.2-Fun-A14B-InP)|`input_image`, `end_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.2-Fun-A14B-InP.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.2-Fun-A14B-InP.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.2-Fun-A14B-InP.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.2-Fun-A14B-InP.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.2-Fun-A14B-InP.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.2-Fun-A14B-InP.py)|
-|[PAI/Wan2.2-Fun-A14B-Control](https://modelscope.cn/models/PAI/Wan2.2-Fun-A14B-Control)|`control_video`, `reference_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.2-Fun-A14B-Control.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.2-Fun-A14B-Control.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.2-Fun-A14B-Control.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.2-Fun-A14B-Control.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.2-Fun-A14B-Control.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.2-Fun-A14B-Control.py)|
-|[PAI/Wan2.2-Fun-A14B-Control-Camera](https://modelscope.cn/models/PAI/Wan2.2-Fun-A14B-Control-Camera)|`control_camera_video`, `input_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.2-Fun-A14B-Control-Camera.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.2-Fun-A14B-Control-Camera.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.2-Fun-A14B-Control-Camera.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.2-Fun-A14B-Control-Camera.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.2-Fun-A14B-Control-Camera.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.2-Fun-A14B-Control-Camera.py)|
-|[openmoss/MOVA-360p](https://modelscope.cn/models/openmoss/MOVA-360p)|`input_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/mova/model_inference/MOVA-360p-I2AV.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/mova/model_inference_low_vram/MOVA-360p-I2AV.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/mova/model_training/full/MOVA-360P-I2AV.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/mova/model_training/validate_full/MOVA-360p-I2AV.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/mova/model_training/lora/MOVA-360P-I2AV.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/mova/model_training/validate_lora/MOVA-360p-I2AV.py)|
-|[openmoss/MOVA-720p](https://modelscope.cn/models/openmoss/MOVA-720p)|`input_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/mova/model_inference/MOVA-720p-I2AV.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/mova/model_inference_low_vram/MOVA-720p-I2AV.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/mova/model_training/full/MOVA-720P-I2AV.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/mova/model_training/validate_full/MOVA-720p-I2AV.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/mova/model_training/lora/MOVA-720P-I2AV.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/mova/model_training/validate_lora/MOVA-720p-I2AV.py)|
-|[Wan-AI/Wan-Dancer-14B (global model)](https://modelscope.cn/models/Wan-AI/Wan-Dancer-14B)|`wantodance_music_path`, `wantodance_reference_image`, `wantodance_fps`, `wantodance_keyframes`, `wantodance_keyframes_mask`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan-Dancer-14B-global.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan-Dancer-14B-global.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan-Dancer-14B-global.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan-Dancer-14B-global.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan-Dancer-14B-global.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan-Dancer-14B-global.py)|
-|[Wan-AI/Wan-Dancer-14B (local model)](https://modelscope.cn/models/Wan-AI/Wan-Dancer-14B)|`wantodance_music_path`, `wantodance_reference_image`, `wantodance_fps`, `wantodance_keyframes`, `wantodance_keyframes_mask`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan-Dancer-14B-local.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan-Dancer-14B-local.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan-Dancer-14B-local.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan-Dancer-14B-local.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan-Dancer-14B-local.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan-Dancer-14B-local.py)|
+|[Wan-AI/Wan2.1-T2V-1.3B](https://modelscope.cn/models/Wan-AI/Wan2.1-T2V-1.3B)||[code](/examples/wanvideo/model_inference/Wan2.1-T2V-1.3B.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.1-T2V-1.3B.py)|[code](/examples/wanvideo/model_training/full/Wan2.1-T2V-1.3B.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.1-T2V-1.3B.py)|[code](/examples/wanvideo/model_training/lora/Wan2.1-T2V-1.3B.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.1-T2V-1.3B.py)|
+|[Wan-AI/Wan2.1-T2V-14B](https://modelscope.cn/models/Wan-AI/Wan2.1-T2V-14B)||[code](/examples/wanvideo/model_inference/Wan2.1-T2V-14B.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.1-T2V-14B.py)|[code](/examples/wanvideo/model_training/full/Wan2.1-T2V-14B.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.1-T2V-14B.py)|[code](/examples/wanvideo/model_training/lora/Wan2.1-T2V-14B.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.1-T2V-14B.py)|
+|[Wan-AI/Wan2.1-I2V-14B-480P](https://modelscope.cn/models/Wan-AI/Wan2.1-I2V-14B-480P)|`input_image`|[code](/examples/wanvideo/model_inference/Wan2.1-I2V-14B-480P.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.1-I2V-14B-480P.py)|[code](/examples/wanvideo/model_training/full/Wan2.1-I2V-14B-480P.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.1-I2V-14B-480P.py)|[code](/examples/wanvideo/model_training/lora/Wan2.1-I2V-14B-480P.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.1-I2V-14B-480P.py)|
+|[Wan-AI/Wan2.1-I2V-14B-720P](https://modelscope.cn/models/Wan-AI/Wan2.1-I2V-14B-720P)|`input_image`|[code](/examples/wanvideo/model_inference/Wan2.1-I2V-14B-720P.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.1-I2V-14B-720P.py)|[code](/examples/wanvideo/model_training/full/Wan2.1-I2V-14B-720P.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.1-I2V-14B-720P.py)|[code](/examples/wanvideo/model_training/lora/Wan2.1-I2V-14B-720P.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.1-I2V-14B-720P.py)|
+|[Wan-AI/Wan2.1-FLF2V-14B-720P](https://modelscope.cn/models/Wan-AI/Wan2.1-FLF2V-14B-720P)|`input_image`, `end_image`|[code](/examples/wanvideo/model_inference/Wan2.1-FLF2V-14B-720P.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.1-FLF2V-14B-720P.py)|[code](/examples/wanvideo/model_training/full/Wan2.1-FLF2V-14B-720P.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.1-FLF2V-14B-720P.py)|[code](/examples/wanvideo/model_training/lora/Wan2.1-FLF2V-14B-720P.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.1-FLF2V-14B-720P.py)|
+|[iic/VACE-Wan2.1-1.3B-Preview](https://modelscope.cn/models/iic/VACE-Wan2.1-1.3B-Preview)|`vace_control_video`, `vace_reference_image`|[code](/examples/wanvideo/model_inference/Wan2.1-VACE-1.3B-Preview.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.1-VACE-1.3B-Preview.py)|[code](/examples/wanvideo/model_training/full/Wan2.1-VACE-1.3B-Preview.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.1-VACE-1.3B-Preview.py)|[code](/examples/wanvideo/model_training/lora/Wan2.1-VACE-1.3B-Preview.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.1-VACE-1.3B-Preview.py)|
+|[Wan-AI/Wan2.1-VACE-1.3B](https://modelscope.cn/models/Wan-AI/Wan2.1-VACE-1.3B)|`vace_control_video`, `vace_reference_image`|[code](/examples/wanvideo/model_inference/Wan2.1-VACE-1.3B.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.1-VACE-1.3B.py)|[code](/examples/wanvideo/model_training/full/Wan2.1-VACE-1.3B.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.1-VACE-1.3B.py)|[code](/examples/wanvideo/model_training/lora/Wan2.1-VACE-1.3B.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.1-VACE-1.3B.py)|
+|[Wan-AI/Wan2.1-VACE-14B](https://modelscope.cn/models/Wan-AI/Wan2.1-VACE-14B)|`vace_control_video`, `vace_reference_image`|[code](/examples/wanvideo/model_inference/Wan2.1-VACE-14B.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.1-VACE-14B.py)|[code](/examples/wanvideo/model_training/full/Wan2.1-VACE-14B.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.1-VACE-14B.py)|[code](/examples/wanvideo/model_training/lora/Wan2.1-VACE-14B.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.1-VACE-14B.py)|
+|[PAI/Wan2.1-Fun-1.3B-InP](https://modelscope.cn/models/PAI/Wan2.1-Fun-1.3B-InP)|`input_image`, `end_image`|[code](/examples/wanvideo/model_inference/Wan2.1-Fun-1.3B-InP.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.1-Fun-1.3B-InP.py)|[code](/examples/wanvideo/model_training/full/Wan2.1-Fun-1.3B-InP.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.1-Fun-1.3B-InP.py)|[code](/examples/wanvideo/model_training/lora/Wan2.1-Fun-1.3B-InP.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.1-Fun-1.3B-InP.py)|
+|[PAI/Wan2.1-Fun-1.3B-Control](https://modelscope.cn/models/PAI/Wan2.1-Fun-1.3B-Control)|`control_video`|[code](/examples/wanvideo/model_inference/Wan2.1-Fun-1.3B-Control.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.1-Fun-1.3B-Control.py)|[code](/examples/wanvideo/model_training/full/Wan2.1-Fun-1.3B-Control.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.1-Fun-1.3B-Control.py)|[code](/examples/wanvideo/model_training/lora/Wan2.1-Fun-1.3B-Control.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.1-Fun-1.3B-Control.py)|
+|[PAI/Wan2.1-Fun-14B-InP](https://modelscope.cn/models/PAI/Wan2.1-Fun-14B-InP)|`input_image`, `end_image`|[code](/examples/wanvideo/model_inference/Wan2.1-Fun-14B-InP.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.1-Fun-14B-InP.py)|[code](/examples/wanvideo/model_training/full/Wan2.1-Fun-14B-InP.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.1-Fun-14B-InP.py)|[code](/examples/wanvideo/model_training/lora/Wan2.1-Fun-14B-InP.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.1-Fun-14B-InP.py)|
+|[PAI/Wan2.1-Fun-14B-Control](https://modelscope.cn/models/PAI/Wan2.1-Fun-14B-Control)|`control_video`|[code](/examples/wanvideo/model_inference/Wan2.1-Fun-14B-Control.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.1-Fun-14B-Control.py)|[code](/examples/wanvideo/model_training/full/Wan2.1-Fun-14B-Control.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.1-Fun-14B-Control.py)|[code](/examples/wanvideo/model_training/lora/Wan2.1-Fun-14B-Control.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.1-Fun-14B-Control.py)|
+|[PAI/Wan2.1-Fun-V1.1-1.3B-Control](https://modelscope.cn/models/PAI/Wan2.1-Fun-V1.1-1.3B-Control)|`control_video`, `reference_image`|[code](/examples/wanvideo/model_inference/Wan2.1-Fun-V1.1-1.3B-Control.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.1-Fun-V1.1-1.3B-Control.py)|[code](/examples/wanvideo/model_training/full/Wan2.1-Fun-V1.1-1.3B-Control.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.1-Fun-V1.1-1.3B-Control.py)|[code](/examples/wanvideo/model_training/lora/Wan2.1-Fun-V1.1-1.3B-Control.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.1-Fun-V1.1-1.3B-Control.py)|
+|[PAI/Wan2.1-Fun-V1.1-14B-Control](https://modelscope.cn/models/PAI/Wan2.1-Fun-V1.1-14B-Control)|`control_video`, `reference_image`|[code](/examples/wanvideo/model_inference/Wan2.1-Fun-V1.1-14B-Control.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.1-Fun-V1.1-14B-Control.py)|[code](/examples/wanvideo/model_training/full/Wan2.1-Fun-V1.1-14B-Control.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.1-Fun-V1.1-14B-Control.py)|[code](/examples/wanvideo/model_training/lora/Wan2.1-Fun-V1.1-14B-Control.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.1-Fun-V1.1-14B-Control.py)|
+|[PAI/Wan2.1-Fun-V1.1-1.3B-InP](https://modelscope.cn/models/PAI/Wan2.1-Fun-V1.1-1.3B-InP)|`input_image`, `end_image`|[code](/examples/wanvideo/model_inference/Wan2.1-Fun-V1.1-1.3B-InP.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.1-Fun-V1.1-1.3B-InP.py)|[code](/examples/wanvideo/model_training/full/Wan2.1-Fun-V1.1-1.3B-InP.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.1-Fun-V1.1-1.3B-InP.py)|[code](/examples/wanvideo/model_training/lora/Wan2.1-Fun-V1.1-1.3B-InP.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.1-Fun-V1.1-1.3B-InP.py)|
+|[PAI/Wan2.1-Fun-V1.1-14B-InP](https://modelscope.cn/models/PAI/Wan2.1-Fun-V1.1-14B-InP)|`input_image`, `end_image`|[code](/examples/wanvideo/model_inference/Wan2.1-Fun-V1.1-14B-InP.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.1-Fun-V1.1-14B-InP.py)|[code](/examples/wanvideo/model_training/full/Wan2.1-Fun-V1.1-14B-InP.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.1-Fun-V1.1-14B-InP.py)|[code](/examples/wanvideo/model_training/lora/Wan2.1-Fun-V1.1-14B-InP.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.1-Fun-V1.1-14B-InP.py)|
+|[PAI/Wan2.1-Fun-V1.1-1.3B-Control-Camera](https://modelscope.cn/models/PAI/Wan2.1-Fun-V1.1-1.3B-Control-Camera)|`control_camera_video`, `input_image`|[code](/examples/wanvideo/model_inference/Wan2.1-Fun-V1.1-1.3B-Control-Camera.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.1-Fun-V1.1-1.3B-Control-Camera.py)|[code](/examples/wanvideo/model_training/full/Wan2.1-Fun-V1.1-1.3B-Control-Camera.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.1-Fun-V1.1-1.3B-Control-Camera.py)|[code](/examples/wanvideo/model_training/lora/Wan2.1-Fun-V1.1-1.3B-Control-Camera.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.1-Fun-V1.1-1.3B-Control-Camera.py)|
+|[PAI/Wan2.1-Fun-V1.1-14B-Control-Camera](https://modelscope.cn/models/PAI/Wan2.1-Fun-V1.1-14B-Control-Camera)|`control_camera_video`, `input_image`|[code](/examples/wanvideo/model_inference/Wan2.1-Fun-V1.1-14B-Control-Camera.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.1-Fun-V1.1-14B-Control-Camera.py)|[code](/examples/wanvideo/model_training/full/Wan2.1-Fun-V1.1-14B-Control-Camera.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.1-Fun-V1.1-14B-Control-Camera.py)|[code](/examples/wanvideo/model_training/lora/Wan2.1-Fun-V1.1-14B-Control-Camera.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.1-Fun-V1.1-14B-Control-Camera.py)|
+|[DiffSynth-Studio/Wan2.1-1.3b-speedcontrol-v1](https://modelscope.cn/models/DiffSynth-Studio/Wan2.1-1.3b-speedcontrol-v1)|`motion_bucket_id`|[code](/examples/wanvideo/model_inference/Wan2.1-1.3b-speedcontrol-v1.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.1-1.3b-speedcontrol-v1.py)|[code](/examples/wanvideo/model_training/full/Wan2.1-1.3b-speedcontrol-v1.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.1-1.3b-speedcontrol-v1.py)|[code](/examples/wanvideo/model_training/lora/Wan2.1-1.3b-speedcontrol-v1.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.1-1.3b-speedcontrol-v1.py)|
+|[krea/krea-realtime-video](https://www.modelscope.cn/models/krea/krea-realtime-video)||[code](/examples/wanvideo/model_inference/krea-realtime-video.py)|[code](/examples/wanvideo/model_inference_low_vram/krea-realtime-video.py)|[code](/examples/wanvideo/model_training/full/krea-realtime-video.sh)|[code](/examples/wanvideo/model_training/validate_full/krea-realtime-video.py)|[code](/examples/wanvideo/model_training/lora/krea-realtime-video.sh)|[code](/examples/wanvideo/model_training/validate_lora/krea-realtime-video.py)|
+|[meituan-longcat/LongCat-Video](https://www.modelscope.cn/models/meituan-longcat/LongCat-Video)|`longcat_video`|[code](/examples/wanvideo/model_inference/LongCat-Video.py)|[code](/examples/wanvideo/model_inference_low_vram/LongCat-Video.py)|[code](/examples/wanvideo/model_training/full/LongCat-Video.sh)|[code](/examples/wanvideo/model_training/validate_full/LongCat-Video.py)|[code](/examples/wanvideo/model_training/lora/LongCat-Video.sh)|[code](/examples/wanvideo/model_training/validate_lora/LongCat-Video.py)|
+|[ByteDance/Video-As-Prompt-Wan2.1-14B](https://modelscope.cn/models/ByteDance/Video-As-Prompt-Wan2.1-14B)|`vap_video`, `vap_prompt`|[code](/examples/wanvideo/model_inference/Video-As-Prompt-Wan2.1-14B.py)|[code](/examples/wanvideo/model_inference_low_vram/Video-As-Prompt-Wan2.1-14B.py)|[code](/examples/wanvideo/model_training/full/Video-As-Prompt-Wan2.1-14B.sh)|[code](/examples/wanvideo/model_training/validate_full/Video-As-Prompt-Wan2.1-14B.py)|[code](/examples/wanvideo/model_training/lora/Video-As-Prompt-Wan2.1-14B.sh)|[code](/examples/wanvideo/model_training/validate_lora/Video-As-Prompt-Wan2.1-14B.py)|
+|[Wan-AI/Wan2.2-T2V-A14B](https://modelscope.cn/models/Wan-AI/Wan2.2-T2V-A14B)||[code](/examples/wanvideo/model_inference/Wan2.2-T2V-A14B.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.2-T2V-A14B.py)|[code](/examples/wanvideo/model_training/full/Wan2.2-T2V-A14B.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.2-T2V-A14B.py)|[code](/examples/wanvideo/model_training/lora/Wan2.2-T2V-A14B.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.2-T2V-A14B.py)|
+|[Wan-AI/Wan2.2-I2V-A14B](https://modelscope.cn/models/Wan-AI/Wan2.2-I2V-A14B)|`input_image`|[code](/examples/wanvideo/model_inference/Wan2.2-I2V-A14B.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.2-I2V-A14B.py)|[code](/examples/wanvideo/model_training/full/Wan2.2-I2V-A14B.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.2-I2V-A14B.py)|[code](/examples/wanvideo/model_training/lora/Wan2.2-I2V-A14B.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.2-I2V-A14B.py)|
+|[Wan-AI/Wan2.2-TI2V-5B](https://modelscope.cn/models/Wan-AI/Wan2.2-TI2V-5B)|`input_image`|[code](/examples/wanvideo/model_inference/Wan2.2-TI2V-5B.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.2-TI2V-5B.py)|[code](/examples/wanvideo/model_training/full/Wan2.2-TI2V-5B.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.2-TI2V-5B.py)|[code](/examples/wanvideo/model_training/lora/Wan2.2-TI2V-5B.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.2-TI2V-5B.py)|
+|[Wan-AI/Wan2.2-Animate-14B](https://www.modelscope.cn/models/Wan-AI/Wan2.2-Animate-14B)|`input_image`, `animate_pose_video`, `animate_face_video`, `animate_inpaint_video`, `animate_mask_video`|[code](/examples/wanvideo/model_inference/Wan2.2-Animate-14B.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.2-Animate-14B.py)|[code](/examples/wanvideo/model_training/full/Wan2.2-Animate-14B.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.2-Animate-14B.py)|[code](/examples/wanvideo/model_training/lora/Wan2.2-Animate-14B.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.2-Animate-14B.py)|
+|[Wan-AI/Wan2.2-Animate-2-14B](https://www.modelscope.cn/models/Wan-AI/Wan2.2-Animate-2-14B)|`animate2_reference_image`, `animate2_reference_video`, `animate2_prompt_ref`|[code](/examples/wanvideo/model_inference/Wan2.2-Animate-2-14B.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.2-Animate-2-14B.py)|[code](/examples/wanvideo/model_training/full/Wan2.2-Animate-2-14B.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.2-Animate-2-14B.py)|[code](/examples/wanvideo/model_training/lora/Wan2.2-Animate-2-14B.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.2-Animate-2-14B.py)|
+|[Wan-AI/Wan2.2-Animate-2-14B: Distilled](https://www.modelscope.cn/models/Wan-AI/Wan2.2-Animate-2-14B)|`animate2_reference_image`, `animate2_reference_video`, `animate2_prompt_ref`|[code](/examples/wanvideo/model_inference/Wan2.2-Animate-2-14B-Distilled.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.2-Animate-2-14B-Distilled.py)|[code](/examples/wanvideo/model_training/full/Wan2.2-Animate-2-14B-Distilled.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.2-Animate-2-14B-Distilled.py)|[code](/examples/wanvideo/model_training/lora/Wan2.2-Animate-2-14B-Distilled.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.2-Animate-2-14B-Distilled.py)|
+|[Wan-AI/Wan2.2-S2V-14B](https://www.modelscope.cn/models/Wan-AI/Wan2.2-S2V-14B)|`input_image`, `input_audio`, `audio_sample_rate`, `s2v_pose_video`|[code](/examples/wanvideo/model_inference/Wan2.2-S2V-14B_multi_clips.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.2-S2V-14B_multi_clips.py)|[code](/examples/wanvideo/model_training/full/Wan2.2-S2V-14B.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.2-S2V-14B.py)|[code](/examples/wanvideo/model_training/lora/Wan2.2-S2V-14B.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.2-S2V-14B.py)|
+|[PAI/Wan2.2-VACE-Fun-A14B](https://www.modelscope.cn/models/PAI/Wan2.2-VACE-Fun-A14B)|`vace_control_video`, `vace_reference_image`|[code](/examples/wanvideo/model_inference/Wan2.2-VACE-Fun-A14B.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.2-VACE-Fun-A14B.py)|[code](/examples/wanvideo/model_training/full/Wan2.2-VACE-Fun-A14B.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.2-VACE-Fun-A14B.py)|[code](/examples/wanvideo/model_training/lora/Wan2.2-VACE-Fun-A14B.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.2-VACE-Fun-A14B.py)|
+|[PAI/Wan2.2-Fun-A14B-InP](https://modelscope.cn/models/PAI/Wan2.2-Fun-A14B-InP)|`input_image`, `end_image`|[code](/examples/wanvideo/model_inference/Wan2.2-Fun-A14B-InP.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.2-Fun-A14B-InP.py)|[code](/examples/wanvideo/model_training/full/Wan2.2-Fun-A14B-InP.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.2-Fun-A14B-InP.py)|[code](/examples/wanvideo/model_training/lora/Wan2.2-Fun-A14B-InP.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.2-Fun-A14B-InP.py)|
+|[PAI/Wan2.2-Fun-A14B-Control](https://modelscope.cn/models/PAI/Wan2.2-Fun-A14B-Control)|`control_video`, `reference_image`|[code](/examples/wanvideo/model_inference/Wan2.2-Fun-A14B-Control.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.2-Fun-A14B-Control.py)|[code](/examples/wanvideo/model_training/full/Wan2.2-Fun-A14B-Control.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.2-Fun-A14B-Control.py)|[code](/examples/wanvideo/model_training/lora/Wan2.2-Fun-A14B-Control.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.2-Fun-A14B-Control.py)|
+|[PAI/Wan2.2-Fun-A14B-Control-Camera](https://modelscope.cn/models/PAI/Wan2.2-Fun-A14B-Control-Camera)|`control_camera_video`, `input_image`|[code](/examples/wanvideo/model_inference/Wan2.2-Fun-A14B-Control-Camera.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan2.2-Fun-A14B-Control-Camera.py)|[code](/examples/wanvideo/model_training/full/Wan2.2-Fun-A14B-Control-Camera.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan2.2-Fun-A14B-Control-Camera.py)|[code](/examples/wanvideo/model_training/lora/Wan2.2-Fun-A14B-Control-Camera.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan2.2-Fun-A14B-Control-Camera.py)|
+|[openmoss/MOVA-360p](https://modelscope.cn/models/openmoss/MOVA-360p)|`input_image`|[code](/examples/mova/model_inference/MOVA-360p-I2AV.py)|[code](/examples/mova/model_inference_low_vram/MOVA-360p-I2AV.py)|[code](/examples/mova/model_training/full/MOVA-360P-I2AV.sh)|[code](/examples/mova/model_training/validate_full/MOVA-360p-I2AV.py)|[code](/examples/mova/model_training/lora/MOVA-360P-I2AV.sh)|[code](/examples/mova/model_training/validate_lora/MOVA-360p-I2AV.py)|
+|[openmoss/MOVA-720p](https://modelscope.cn/models/openmoss/MOVA-720p)|`input_image`|[code](/examples/mova/model_inference/MOVA-720p-I2AV.py)|[code](/examples/mova/model_inference_low_vram/MOVA-720p-I2AV.py)|[code](/examples/mova/model_training/full/MOVA-720P-I2AV.sh)|[code](/examples/mova/model_training/validate_full/MOVA-720p-I2AV.py)|[code](/examples/mova/model_training/lora/MOVA-720P-I2AV.sh)|[code](/examples/mova/model_training/validate_lora/MOVA-720p-I2AV.py)|
+|[Wan-AI/Wan-Dancer-14B (global model)](https://modelscope.cn/models/Wan-AI/Wan-Dancer-14B)|`wantodance_music_path`, `wantodance_reference_image`, `wantodance_fps`, `wantodance_keyframes`, `wantodance_keyframes_mask`|[code](/examples/wanvideo/model_inference/Wan-Dancer-14B-global.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan-Dancer-14B-global.py)|[code](/examples/wanvideo/model_training/full/Wan-Dancer-14B-global.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan-Dancer-14B-global.py)|[code](/examples/wanvideo/model_training/lora/Wan-Dancer-14B-global.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan-Dancer-14B-global.py)|
+|[Wan-AI/Wan-Dancer-14B (local model)](https://modelscope.cn/models/Wan-AI/Wan-Dancer-14B)|`wantodance_music_path`, `wantodance_reference_image`, `wantodance_fps`, `wantodance_keyframes`, `wantodance_keyframes_mask`|[code](/examples/wanvideo/model_inference/Wan-Dancer-14B-local.py)|[code](/examples/wanvideo/model_inference_low_vram/Wan-Dancer-14B-local.py)|[code](/examples/wanvideo/model_training/full/Wan-Dancer-14B-local.sh)|[code](/examples/wanvideo/model_training/validate_full/Wan-Dancer-14B-local.py)|[code](/examples/wanvideo/model_training/lora/Wan-Dancer-14B-local.sh)|[code](/examples/wanvideo/model_training/validate_lora/Wan-Dancer-14B-local.py)|
 
 </details>
 
diff --git a/diffsynth/configs/model_configs.py b/diffsynth/configs/model_configs.py
index 2b736e5a4..437c68179 100644
--- a/diffsynth/configs/model_configs.py
+++ b/diffsynth/configs/model_configs.py
@@ -81,6 +81,13 @@
 ]
 
 wan_series = [
+    {
+        # Example: ModelConfig(model_id="Wan-AI/Wan2.2-Animate-2-14B", origin_file_pattern="wan_animate_2/wan_animate_2_bf16.safetensors")
+        "model_hash": "4536c21ad8740ba78367af4216ae85bf",
+        "model_name": "wan_video_dit",
+        "model_class": "diffsynth.models.wan_animate_2_dit.WanAnimate2Transformer",
+        "extra_kwargs": {},
+    },
     {
         # Example: ModelConfig(model_id="krea/krea-realtime-video", origin_file_pattern="krea-realtime-video-14b.safetensors")
         "model_hash": "5ec04e02b42d2580483ad69f4e76346a",
diff --git a/diffsynth/configs/vram_management_module_maps.py b/diffsynth/configs/vram_management_module_maps.py
index 8d83b0fa9..4f874756e 100644
--- a/diffsynth/configs/vram_management_module_maps.py
+++ b/diffsynth/configs/vram_management_module_maps.py
@@ -114,6 +114,15 @@
         "diffsynth.models.wan_video_dit.RMSNorm": "diffsynth.core.vram.layers.AutoWrappedModule",
         "torch.nn.Conv2d": "diffsynth.core.vram.layers.AutoWrappedModule",
     },
+    "diffsynth.models.wan_animate_2_dit.WanAnimate2Transformer": {
+        "diffsynth.models.wan_video_dit.MLP": "diffsynth.core.vram.layers.AutoWrappedModule",
+        "diffsynth.models.wan_animate_2_dit.AttentionBlock": "diffsynth.core.vram.layers.AutoWrappedNonRecurseModule",
+        "diffsynth.models.wan_animate_2_dit.Head": "diffsynth.core.vram.layers.AutoWrappedModule",
+        "torch.nn.Linear": "diffsynth.core.vram.layers.AutoWrappedLinear",
+        "torch.nn.Conv3d": "diffsynth.core.vram.layers.AutoWrappedModule",
+        "torch.nn.LayerNorm": "diffsynth.core.vram.layers.AutoWrappedModule",
+        "diffsynth.models.wan_video_dit.RMSNorm": "diffsynth.core.vram.layers.AutoWrappedModule",
+    },
     "diffsynth.models.wan_video_dit.WanModel": {
         "diffsynth.models.wan_video_dit.MLP": "diffsynth.core.vram.layers.AutoWrappedModule",
         "diffsynth.models.wan_video_dit.DiTBlock": "diffsynth.core.vram.layers.AutoWrappedNonRecurseModule",
diff --git a/diffsynth/core/attention/attention.py b/diffsynth/core/attention/attention.py
index 642424cd3..7cff7c14a 100644
--- a/diffsynth/core/attention/attention.py
+++ b/diffsynth/core/attention/attention.py
@@ -26,6 +26,13 @@
 except ModuleNotFoundError:
     XFORMERS_AVAILABLE = False
 
+try:
+    from torch.nn.attention.flex_attention import flex_attention as flex_attention_func
+    flex_attention_func = torch.compile(flex_attention_func, dynamic=False, mode="max-autotune-no-cudagraphs", fullgraph=True, backend="inductor")
+    FLEX_ATTN_AVAILABLE = True
+except (ModuleNotFoundError, ImportError):
+    FLEX_ATTN_AVAILABLE = False
+
 try:
     if "enable_gqa" in inspect.signature(torch.nn.functional.scaled_dot_product_attention).parameters:
         TORCH_SUPPORT_GQA = True
@@ -169,9 +176,20 @@ def xformers_attention(q: torch.Tensor, k: torch.Tensor, v: torch.Tensor, q_patt
     return out
 
 
-def attention_forward(q: torch.Tensor, k: torch.Tensor, v: torch.Tensor, q_pattern="b n s d", k_pattern="b n s d", v_pattern="b n s d", out_pattern="b n s d", dims=None, attn_mask=None, scale=None, is_causal=False, compatibility_mode=False, window_size=None):
+def flex_attention(q: torch.Tensor, k: torch.Tensor, v: torch.Tensor, q_pattern="b n s d", k_pattern="b n s d", v_pattern="b n s d", out_pattern="b n s d", dims=None, attn_mask=None, scale=None, score_mod=None):
+    assert FLEX_ATTN_AVAILABLE, "Flex Attention is not available. Please upgrade torch to 2.5.0 or later."
+    required_in_pattern, required_out_pattern = "b n s d", "b n s d"
+    q, k, v = rearrange_qkv(q, k, v, q_pattern, k_pattern, v_pattern, required_in_pattern, dims)
+    out = flex_attention_func(query=q, key=k, value=v, block_mask=attn_mask, scale=scale, score_mod=score_mod)
+    out = rearrange_out(out, out_pattern, required_out_pattern, dims)
+    return out
+
+
+def attention_forward(q: torch.Tensor, k: torch.Tensor, v: torch.Tensor, q_pattern="b n s d", k_pattern="b n s d", v_pattern="b n s d", out_pattern="b n s d", dims=None, attn_mask=None, scale=None, is_causal=False, compatibility_mode=False, window_size=None, use_flex=False, score_mod=None):
     if compatibility_mode or (attn_mask is not None) or ATTENTION_IMPLEMENTATION == "torch":
-        if window_size is None:
+        if use_flex or score_mod is not None:
+            return flex_attention(q, k, v, q_pattern, k_pattern, v_pattern, out_pattern, dims, attn_mask=attn_mask, scale=scale, score_mod=score_mod)
+        elif window_size is None:
             return torch_sdpa(q, k, v, q_pattern, k_pattern, v_pattern, out_pattern, dims, attn_mask=attn_mask, scale=scale, is_causal=is_causal)
         else:
             # Sliding Window Attention is not compatible with `is_causal` and `attn_mask`.
diff --git a/diffsynth/models/wan_animate_2_dit.py b/diffsynth/models/wan_animate_2_dit.py
new file mode 100644
index 000000000..54750db90
--- /dev/null
+++ b/diffsynth/models/wan_animate_2_dit.py
@@ -0,0 +1,913 @@
+import numpy as np
+import torch
+import torch.nn as nn
+import math
+from torch.nn.attention.flex_attention import create_block_mask
+from .wan_video_dit import sinusoidal_embedding_1d, RMSNorm, MLP
+from ..core.attention.attention import attention_forward
+from ..core.gradient import gradient_checkpoint_forward
+from functools import lru_cache, partial
+
+
+def _score_mod_impl(score, b_idx, h_idx, q_idx, kv_idx, hw: int, log_scale: float):
+    condition = (kv_idx >= hw) & (kv_idx < 2 * hw)
+    return torch.where(condition, score + log_scale, score)
+
+
+@lru_cache(maxsize=32)
+def _get_score_mod(hw: int, log_scale: float = -1.0):
+    return partial(_score_mod_impl, hw=hw, log_scale=log_scale)
+
+
+def rope_params(max_seq_len, dim, theta=10000, offset=0):
+    assert dim % 2 == 0
+    freqs = torch.outer(
+        torch.arange(max_seq_len)+offset,
+        1.0 / torch.pow(theta, torch.arange(0, dim, 2).to(torch.float64).div(dim))
+    )
+    freqs = torch.polar(torch.ones_like(freqs), freqs)
+    return freqs
+
+
+def rope_apply(x, grid_sizes, freqs, time_stride=1):
+    n, c = x.size(2), x.size(3) // 2
+
+    # split freqs
+    freqs = freqs.split([c - 2 * (c // 3), c // 3, c // 3], dim=1)
+
+    # loop over samples
+    output = []
+    for i, (f, h, w) in enumerate(grid_sizes.tolist()):
+        seq_len = f * h * w
+
+        # precompute multipliers
+        x_i = torch.view_as_complex(
+            x[i, :seq_len].to(torch.float64).reshape(seq_len, n, -1, 2)
+        )
+        freqs_i = torch.cat([
+            freqs[0][:f*time_stride:time_stride].view(f, 1, 1, -1).expand(f, h, w, -1),
+            freqs[1][:h].view(1, h, 1, -1).expand(f, h, w, -1),
+            freqs[2][:w].view(1, 1, w, -1).expand(f, h, w, -1)
+        ], dim=-1).reshape(seq_len, 1, -1)
+
+        # apply rotary embedding
+        x_i = torch.view_as_real(x_i * freqs_i).flatten(2)
+        x_i = torch.cat([x_i, x[i, seq_len:]])
+
+        # append to collection
+        output.append(x_i)
+    return torch.stack(output).to(x.dtype)
+
+
+def pad_freqs(original_tensor, target_len):
+    seq_len, s1, s2 = original_tensor.shape
+    pad_size = target_len - seq_len
+    padding_tensor = torch.ones(
+        pad_size,
+        s1,
+        s2,
+        dtype=original_tensor.dtype,
+        device=original_tensor.device)
+    padded_tensor = torch.cat([original_tensor, padding_tensor], dim=0)
+    return padded_tensor
+
+
+class LayerNorm(nn.LayerNorm):
+
+    def __init__(self, dim, eps=1e-6, elementwise_affine=False):
+        super().__init__(dim, elementwise_affine=elementwise_affine, eps=eps)
+
+    def forward(self, x):
+        return super().forward(x).type_as(x)
+
+
+class SelfAttention(nn.Module):
+
+    def __init__(
+        self,
+        dim,
+        num_heads,
+        window_size=(-1, -1),
+        qk_norm=True,
+        eps=1e-6
+    ):
+        assert dim % num_heads == 0
+        super().__init__()
+        self.dim = dim
+        self.num_heads = num_heads
+        self.head_dim = dim // num_heads
+        self.window_size = window_size
+        self.qk_norm = qk_norm
+        self.eps = eps
+
+        # layers
+        self.q = nn.Linear(dim, dim)
+        self.k = nn.Linear(dim, dim)
+        self.v = nn.Linear(dim, dim)
+        self.o = nn.Linear(dim, dim)
+        self.norm_q = RMSNorm(dim, eps=eps) if qk_norm else nn.Identity()
+        self.norm_k = RMSNorm(dim, eps=eps) if qk_norm else nn.Identity()
+
+    def forward(self, *args, method, **kwargs):
+        return getattr(self, method)(*args, **kwargs)
+
+    def pre_attention(self, x):
+        b, s, n, d = *x.shape[:2], self.num_heads, self.head_dim
+
+        # query, key, value function
+        def qkv_fn(x):
+            q = self.norm_q(self.q(x)).view(b, s, n, d)
+            k = self.norm_k(self.k(x)).view(b, s, n, d)
+            v = self.v(x).view(b, s, n, d)
+            return q, k, v
+
+        q, k, v = qkv_fn(x)
+
+        return q, k, v
+
+    def post_attention(self, x):
+        # output
+        x = x.flatten(2)
+        x = self.o(x)
+        return x
+
+
+class CrossAttention(SelfAttention):
+    def __init__(self, dim, num_heads, window_size=(-1, -1), qk_norm=True, eps=1e-6, use_img_emb=True):
+        super().__init__(dim, num_heads, window_size, qk_norm, eps)
+        self.use_img_emb = use_img_emb
+        if use_img_emb:
+            self.k_img = nn.Linear(dim, dim)
+            self.v_img = nn.Linear(dim, dim)
+            self.norm_k_img = RMSNorm(dim, eps=eps) if qk_norm else nn.Identity()
+
+    def forward(self, x, context, context_lens, counter=0):
+        if self.use_img_emb:
+            context_img = context[:, :257]
+            context = context[:, 257:]
+
+        b, n, d = x.size(0), self.num_heads, self.head_dim
+
+        # compute query, key, value
+        q = self.norm_q(self.q(x)).view(b, -1, n, d)
+        k = self.norm_k(self.k(context)).view(b, -1, n, d)
+        v = self.v(context).view(b, -1, n, d)
+
+        if self.use_img_emb:
+            k_img = self.norm_k_img(self.k_img(context_img)).view(b, -1, n, d)
+            v_img = self.v_img(context_img).view(b, -1, n, d)
+            img_x = attention_forward(
+                q, k_img, v_img,
+                q_pattern="b s n d", k_pattern="b s n d", v_pattern="b s n d", out_pattern="b s n d",
+            )
+        # compute attention
+        x = attention_forward(
+            q, k, v,
+            q_pattern="b s n d", k_pattern="b s n d", v_pattern="b s n d", out_pattern="b s n d",
+        )
+
+        # output
+        x = x.flatten(2)
+        if self.use_img_emb:
+            img_x = img_x.flatten(2)
+            x = x + img_x
+        x = self.o(x)
+        return x
+
+
+class AttentionBlock(nn.Module):
+
+    def __init__(
+        self,
+        dim,
+        ffn_dim,
+        num_heads,
+        window_size=(-1, -1),
+        qk_norm=True,
+        cross_attn_norm=False,
+        eps=1e-6,
+        use_img_emb=True
+    ):
+        super().__init__()
+        self.dim = dim
+        self.ffn_dim = ffn_dim
+        self.num_heads = num_heads
+        self.window_size = window_size
+        self.qk_norm = qk_norm
+        self.cross_attn_norm = cross_attn_norm
+        self.eps = eps
+
+        # layers
+        self.norm1 = LayerNorm(dim, eps)
+
+        self.self_attn = SelfAttention(dim, num_heads, window_size, qk_norm, eps)
+
+        self.norm3 = LayerNorm(
+            dim, eps, elementwise_affine=True
+        ) if cross_attn_norm else nn.Identity()
+
+        self.cross_attn = CrossAttention(dim, num_heads, (-1, -1), qk_norm, eps, use_img_emb=use_img_emb)
+
+        self.norm2 = LayerNorm(dim, eps)
+        self.ffn = nn.Sequential(
+            nn.Linear(dim, ffn_dim),
+            nn.GELU(approximate='tanh'),
+            nn.Linear(ffn_dim, dim)
+        )
+        # modulation
+        self.modulation = nn.Parameter(torch.randn(1, 6, dim) / dim ** 0.5)
+
+    def forward(self, *args, method, **kwargs):
+        return getattr(self, method)(*args, **kwargs)
+
+    def pre_self_attention(self, x, e):
+        e = (self.modulation.to(x.device) + e).chunk(6, dim=1)
+        q, k, v = self.self_attn(self.norm1(x) * (1 + e[1]) + e[0], method="pre_attention")
+        return q, k, v, e
+
+    def post_self_attention(self, x):
+        x = self.self_attn(x, method="post_attention")
+        return x
+
+    def cross_attention(self, x, context, context_lens, e):
+        x = x + self.cross_attn(self.norm3(x), context, context_lens)
+        y = self.ffn(self.norm2(x) * (1 + e[4]) + e[3])
+        x = x + y * e[5]
+        return x
+
+
+class Incontext_AttentionBlock(nn.Module):
+    def __init__(
+        self,
+        dim,
+        ffn_dim,
+        num_heads,
+        window_size=(-1, -1),
+        qk_norm=True,
+        cross_attn_norm=False,
+        eps=1e-6,
+        refer_stride=1,
+        use_img_emb=True,
+        use_context_parallel=False,
+        sparse_type=0,
+    ):
+        super().__init__()
+        self.dim = dim
+        self.ffn_dim = ffn_dim
+        self.num_heads = num_heads
+        self.window_size = window_size
+        self.qk_norm = qk_norm
+        self.cross_attn_norm = cross_attn_norm
+        self.eps = eps
+        self.refer_stride = refer_stride
+        self.use_context_parallel = use_context_parallel
+        self.sparse_type = sparse_type
+
+        self.block = AttentionBlock(
+            dim, ffn_dim, num_heads, window_size, qk_norm, cross_attn_norm, eps, use_img_emb=use_img_emb
+        )
+
+    def forward(self, *args, method, **kwargs):
+        return getattr(self, method)(*args, **kwargs)
+
+    def forward_ref(
+        self,
+        x_ref,
+        index,
+        k_cache,
+        v_cache,
+        context_ref,
+        freqs_ref,
+        grid_sizes_ref,
+        e_ref,
+        context_lens,
+        animate2_offload_kv=False,
+        use_context_parallel=False,
+    ):
+        q_ref, k_ref, v_ref, e_ref = self.block(x_ref, e_ref, method='pre_self_attention')
+
+        if use_context_parallel:
+            from ..utils.xfuser import all_to_all_4d, is_evenly_divisible, get_sequence_parallel_world_size
+            assert is_evenly_divisible(q_ref.shape[2]), (
+                f"num_heads ({q_ref.shape[2]}) must be divisible by sequence parallel world size ({get_sequence_parallel_world_size()}) "
+            )
+            qkv_ref = torch.cat([q_ref, k_ref, v_ref], dim=0)
+            qkv_ref = all_to_all_4d(qkv_ref, scatter_dim=2, gather_dim=1)
+            q_ref, k_ref, v_ref = qkv_ref.chunk(3, dim=0)
+
+        k_cache[index] = k_ref if not animate2_offload_kv else k_ref.to('cpu')
+        v_cache[index] = v_ref if not animate2_offload_kv else v_ref.to('cpu')
+        q_ref_add_rope = rope_apply(q_ref, grid_sizes_ref, freqs_ref, self.refer_stride)
+        k_ref_add_rope = rope_apply(k_ref, grid_sizes_ref, freqs_ref, self.refer_stride)
+
+        ref_f, ref_h, ref_w = grid_sizes_ref[0].tolist()
+        ref_vail_len = ref_f * ref_h * ref_w
+
+        xout_ref = attention_forward(
+                q_ref_add_rope.to(v_ref.dtype),
+                k_ref_add_rope[:, :ref_vail_len].to(v_ref.dtype),
+                v_ref[:, :ref_vail_len],
+                q_pattern="b s n d", k_pattern="b s n d", v_pattern="b s n d", out_pattern="b s n d",
+        ).to(q_ref_add_rope.dtype)
+
+        if use_context_parallel:
+            xout_ref = all_to_all_4d(xout_ref, scatter_dim=1, gather_dim=2)
+
+        y_ref = self.block(xout_ref, method='post_self_attention')
+
+        x_ref = x_ref + y_ref * e_ref[2]
+
+        x_ref = self.block(x_ref, context_ref, context_lens, e_ref, method='cross_attention')
+
+        return x_ref
+
+    def forward_gen(
+        self,
+        x,
+        index,
+        k_cache,
+        v_cache,
+        attn_mask,
+        context,
+        freqs,
+        freqs_ref,
+        grid_sizes,
+        grid_sizes_ref,
+        origin_len,
+        origin_area,
+        e,
+        context_lens,
+        use_context_parallel=False,
+        log_scale=0.0,
+    ):
+
+        origin_latent_f    = origin_len // 4 + 1
+        origin_latent_hw   = origin_area[0] * origin_area[1] // 256
+        origin_max_len     = (origin_latent_f + 1) * origin_latent_hw
+        origin_ref_max_len = origin_latent_f * origin_latent_hw
+
+        f, h, w = grid_sizes[0].tolist()
+        vail_len = f * h * w
+        hw = h * w
+
+        ref_f, ref_h, ref_w = grid_sizes_ref[0].tolist()
+        ref_vail_len = ref_f * ref_h * ref_w
+        ref_hw = ref_h * ref_w
+
+        q, k, v, e = self.block(x, e, method='pre_self_attention')
+        if use_context_parallel:
+            from ..utils.xfuser import all_to_all_4d, is_evenly_divisible, get_sequence_parallel_world_size
+            assert is_evenly_divisible(q.shape[2]), (
+                f"num_heads ({q.shape[2]}) must be divisible by sequence parallel world size ({get_sequence_parallel_world_size()}) "
+            )
+            qkv = torch.cat([q, k, v], dim=0)
+            qkv = all_to_all_4d(qkv, scatter_dim=2, gather_dim=1)
+            q, k, v = qkv.chunk(3, dim=0)
+
+        q = rope_apply(q, grid_sizes, freqs)
+        k = rope_apply(k, grid_sizes, freqs)
+        k_ref, v_ref = k_cache[index].to(x.device), v_cache[index].to(x.device)
+        k_ref = rope_apply(k_ref, grid_sizes_ref, freqs_ref, self.refer_stride)
+
+        B, _, N, C = q.shape
+        device, dtype = q.device, q.dtype
+
+        target_q_len   = math.ceil(origin_max_len     / 128) * 128
+        target_ref_len = math.ceil(origin_ref_max_len / 128) * 128
+        target_kv_len  = target_q_len + target_ref_len
+
+        q_padding = q[:, vail_len:].clone()
+
+        q_incontext = torch.zeros(B, target_q_len,  N, C, device=device, dtype=dtype)
+        k_incontext = torch.zeros(B, target_kv_len, N, C, device=device, dtype=dtype)
+        v_incontext = torch.zeros(B, target_kv_len, N, C, device=device, dtype=dtype)
+
+        q_src = q[:, :vail_len].view(B, f, hw, N, C)
+        k_src = k[:, :vail_len].view(B, f, hw, N, C)
+        v_src = v[:, :vail_len].view(B, f, hw, N, C)
+
+        q_incontext[:, :f * origin_latent_hw].view(B, f, origin_latent_hw, N, C)[:, :, :hw] = q_src
+        k_incontext[:, :f * origin_latent_hw].view(B, f, origin_latent_hw, N, C)[:, :, :hw] = k_src
+        v_incontext[:, :f * origin_latent_hw].view(B, f, origin_latent_hw, N, C)[:, :, :hw] = v_src
+
+        k_ref_src = k_ref[:, :ref_vail_len].view(B, ref_f, ref_hw, N, C)
+        v_ref_src = v_ref[:, :ref_vail_len].view(B, ref_f, ref_hw, N, C)
+
+        k_incontext[:, target_q_len : target_q_len + ref_f * origin_latent_hw]\
+            .view(B, ref_f, origin_latent_hw, N, C)[:, :, :ref_hw] = k_ref_src
+        v_incontext[:, target_q_len : target_q_len + ref_f * origin_latent_hw]\
+            .view(B, ref_f, origin_latent_hw, N, C)[:, :, :ref_hw] = v_ref_src
+
+        score_mod = _get_score_mod(hw=int(origin_latent_hw), log_scale=log_scale)
+
+        xout_full = attention_forward(
+            q_incontext, k_incontext, v_incontext,
+            q_pattern="b s n d", k_pattern="b s n d", v_pattern="b s n d", out_pattern="b s n d",
+            use_flex=True, attn_mask=attn_mask, score_mod=score_mod
+        )
+
+        xout_valid = xout_full[:, :f * origin_latent_hw]
+        xout_valid = xout_valid.view(B, f, origin_latent_hw, N, C)
+        xout_vail = xout_valid[:, :, :hw]
+        xout_vail = xout_vail.reshape(B, f * hw, N, C)  # [B, f*hw, N, C]
+        xout = torch.cat([xout_vail, q_padding], dim=1)
+        if use_context_parallel:
+            xout = all_to_all_4d(xout, scatter_dim=1, gather_dim=2)
+
+        y = self.block(xout, method='post_self_attention')
+
+        x = x + y * e[2]
+
+        x = self.block(x, context, context_lens, e, method='cross_attention')
+        return x
+
+    def forward_origin(self, x, x_ref, ref_args, gen_args):
+        k_cache, v_cache = {}, {}
+        x_ref = self.forward_ref(x_ref, 0, k_cache, v_cache, **ref_args)
+        x = self.forward_gen(x, 0, k_cache, v_cache, **gen_args)
+        return x, x_ref
+
+
+class Head(nn.Module):
+
+    def __init__(self, dim, out_dim, patch_size, eps=1e-6):
+        super().__init__()
+        self.dim = dim
+        self.out_dim = out_dim
+        self.patch_size = patch_size
+        self.eps = eps
+
+        # layers
+        out_dim = math.prod(patch_size) * out_dim
+        self.norm = LayerNorm(dim, eps)
+        self.head = nn.Linear(dim, out_dim)
+
+        # modulation
+        self.modulation = nn.Parameter(torch.randn(1, 2, dim) / dim**0.5)
+
+    def forward(self, x, e):
+        e = (self.modulation + e.unsqueeze(1)).chunk(2, dim=1)
+        x = self.head(self.norm(x) * (1 + e[1]) + e[0])
+        return x
+
+
+class WanAnimate2Transformer(nn.Module):
+
+    def __init__(
+        self,
+        patch_size=(1, 2, 2),
+        text_len=512,
+        in_dim=36,
+        dim=5120,
+        ffn_dim=13824,
+        freq_dim=256,
+        text_dim=4096,
+        out_dim=16,
+        num_heads=40,
+        num_layers=40,
+        window_size=(-1, -1),
+        qk_norm=True,
+        cross_attn_norm=True,
+        eps=1e-6,
+        use_img_emb=True,
+        refer_offset_t=1,
+        refer_offset_h=0,
+        refer_offset_w=-1,
+        refer_stride=1,
+        sparse_type=0,
+    ):
+        super().__init__()
+        self.patch_size = patch_size
+        self.text_len = text_len
+        self.in_dim = in_dim
+        self.dim = dim
+        self.ffn_dim = ffn_dim
+        self.freq_dim = freq_dim
+        self.text_dim = text_dim
+        self.out_dim = out_dim
+        self.num_heads = num_heads
+        self.num_layers = num_layers
+        self.window_size = window_size
+        self.qk_norm = qk_norm
+        self.cross_attn_norm = cross_attn_norm
+        self.eps = eps
+        self.use_img_emb = use_img_emb
+        self.refer_offset_t = refer_offset_t
+        self.refer_offset_h = refer_offset_h
+        self.refer_offset_w = refer_offset_w
+        self.refer_stride = refer_stride
+        self.sparse_type = sparse_type
+
+        # [Denoising Transformer]
+        # embeddings
+        self.patch_embedding = nn.Conv3d(
+            in_dim, dim, kernel_size=patch_size, stride=patch_size
+        )
+        self.text_embedding = nn.Sequential(
+            nn.Linear(text_dim, dim),
+            nn.GELU(approximate='tanh'),
+            nn.Linear(dim, dim)
+        )
+
+        self.time_embedding = nn.Sequential(
+            nn.Linear(freq_dim, dim),
+            nn.SiLU(),
+            nn.Linear(dim, dim)
+        )
+        self.time_projection = nn.Sequential(
+            nn.SiLU(),
+            nn.Linear(dim, dim * 6)
+        )
+
+        # blocks
+        self.blocks = nn.ModuleList([Incontext_AttentionBlock(
+            dim, ffn_dim, num_heads, window_size, qk_norm, cross_attn_norm, eps, refer_stride, use_img_emb=use_img_emb, sparse_type=self.sparse_type
+        ) for _ in range(num_layers)])
+
+        # head
+        self.head = Head(dim, out_dim, patch_size, eps)
+
+        if use_img_emb:
+            self.img_emb = MLP(1280, dim, has_pos_emb=False)
+
+        self.block_masks = dict()
+        self.block_mask_grid_sizes = dict()
+
+    def create_mask(self, origin_len, origin_area, device):
+        origin_latent_f = origin_len // 4 + 1
+        hw = int(np.prod(origin_area).item() // 256)
+
+        q_len = (origin_latent_f + 1) * hw
+        k_len = origin_latent_f * hw
+
+        q_len_total = math.ceil(q_len / 128) * 128
+        k_extra_len_total = math.ceil(k_len / 128) * 128
+        k_len_total = q_len_total + k_extra_len_total
+
+        q_limit = q_len
+        k_limit = k_len
+        q_total = q_len_total
+
+        def attention_mask_logic(b, h, q_idx, kv_idx):
+            q_valid = q_idx < q_limit
+            is_base_attention = kv_idx < q_limit
+
+            q_frame = q_idx // hw
+            is_first_part = kv_idx < q_total
+
+            kv_frame_1 = kv_idx // hw
+            kv_is_valid_1 = kv_idx < q_limit
+
+            rel_kv_idx = kv_idx - q_total
+            kv_frame_2 = (rel_kv_idx // hw) + 1
+            kv_is_valid_2 = rel_kv_idx < k_limit
+
+            kv_frame = torch.where(is_first_part, kv_frame_1, kv_frame_2)
+            kv_is_valid = torch.where(is_first_part, kv_is_valid_1, kv_is_valid_2)
+
+            is_cond_attention = (q_frame == kv_frame) & kv_is_valid
+
+            return q_valid & (is_base_attention | is_cond_attention)
+
+        block_mask = create_block_mask(
+            attention_mask_logic,
+            B=None,
+            H=None,
+            Q_LEN=q_len_total,
+            KV_LEN=k_len_total,
+            device=device,
+            _compile=True
+        )
+        return block_mask
+
+    def forward_ref(
+        self,
+        x_ref,
+        grid_sizes,
+        k_cache,
+        v_cache,
+        clip_fea_ref,
+        y_ref,
+        context_ref,
+        seq_len_ref,
+        t,
+        animate2_offload_kv=False,
+        use_unified_sequence_parallel: bool = False,
+    ):
+        # [reference]
+        x_ref = [torch.cat([u, v], dim=0) for u, v in zip(x_ref, y_ref)]
+        # embeddings
+        x_ref = [self.patch_embedding(u.unsqueeze(0)) for u in x_ref]
+        grid_sizes_ref = torch.stack([
+            torch.tensor(u.shape[2:], dtype=torch.long) for u in x_ref
+        ])
+        x_ref = [u.flatten(2).transpose(1, 2) for u in x_ref]
+        seq_lens_ref = torch.tensor([u.size(1) for u in x_ref], dtype=torch.long)
+        assert seq_lens_ref.max() <= seq_len_ref
+        x_ref = torch.cat([torch.cat([
+            u, u.new_zeros(1, seq_len_ref - u.size(1), u.size(2))
+        ], dim=1) for u in x_ref])
+
+        assert (self.dim % self.num_heads) == 0 and (self.dim // self.num_heads) % 2 == 0
+        d = self.dim // self.num_heads
+
+        if self.refer_offset_t < 0:
+            self.refer_offset_t = grid_sizes[0][0].item()
+        if self.refer_offset_h < 0:
+            self.refer_offset_h = grid_sizes[0][1].item()
+        if self.refer_offset_w < 0:
+            self.refer_offset_w = grid_sizes[0][2].item()
+
+        self.freqs_ref = torch.cat([
+            rope_params(512, d - 4 * (d // 6), offset=self.refer_offset_t),
+            rope_params(512, 2 * (d // 6), offset=self.refer_offset_h),
+            rope_params(512, 2 * (d // 6), offset=self.refer_offset_w)
+        ], dim=1).to(x_ref.device)
+
+        # time embeddings ref
+        e_ref = self.time_embedding(sinusoidal_embedding_1d(self.freq_dim, t*0+1).to(x_ref.dtype))
+        e0_ref = self.time_projection(e_ref).unflatten(1, (6, self.dim))
+
+        # [context_ref]
+        context_ref = self.text_embedding(torch.stack([torch.cat([
+            u, u.new_zeros(self.text_len - u.size(0), u.size(1))
+        ]) for u in context_ref]))
+
+        if self.use_img_emb:
+            context_clip_ref = self.img_emb(clip_fea_ref) # bs x 257 x dim
+            context_ref = torch.concat([context_clip_ref, context_ref], dim=1)
+
+        context_lens = None
+        # arguments
+        kwargs = dict(
+            e_ref=e0_ref,
+            grid_sizes_ref=grid_sizes_ref,
+            freqs_ref=self.freqs_ref,
+            context_ref=context_ref,
+            context_lens=context_lens,
+            animate2_offload_kv=animate2_offload_kv,
+            use_context_parallel=use_unified_sequence_parallel,
+        )
+        if use_unified_sequence_parallel:
+            from ..utils.xfuser import get_current_chunk, is_evenly_divisible, get_sequence_parallel_world_size
+            assert is_evenly_divisible(x_ref.shape[1]), (
+                f"x_ref sequence length ({x_ref.shape[1]}) must be divisible by sequence parallel world size ({get_sequence_parallel_world_size()}) "
+            )
+            x_ref = get_current_chunk(x_ref, dim=1)
+        for idx, block in enumerate(self.blocks):
+            x_ref = block(x_ref, idx, k_cache, v_cache, method='forward_ref', **kwargs)
+
+    def forward_gen(
+        self,
+        x,
+        k_cache,
+        v_cache,
+        clip_fea,
+        y,
+        context,
+        seq_len,
+        t,
+        grid_sizes_ref,
+        origin_len,
+        origin_area,
+        is_uncondtion=False,
+        use_unified_sequence_parallel: bool = False,
+        log_scale=0.0,
+    ):
+        # params
+        x = [torch.cat([u, v], dim=0) for u, v in zip(x, y)]
+        # embeddings
+        x = [self.patch_embedding(u.unsqueeze(0)) for u in x]
+        grid_sizes = torch.stack([
+            torch.tensor(u.shape[2:], dtype=torch.long) for u in x
+        ])
+        x = [u.flatten(2).transpose(1, 2) for u in x]
+        seq_lens = torch.tensor([u.size(1) for u in x], dtype=torch.long)
+        assert seq_lens.max() <= seq_len
+        x = torch.cat([torch.cat([
+            u, u.new_zeros(1, seq_len - u.size(1), u.size(2))
+        ], dim=1) for u in x])
+
+        assert (self.dim % self.num_heads) == 0 and (self.dim // self.num_heads) % 2 == 0
+        d = self.dim // self.num_heads
+        self.freqs = torch.cat([
+            rope_params(512, d - 4 * (d // 6)),
+            rope_params(512, 2 * (d // 6)),
+            rope_params(512, 2 * (d // 6))
+        ], dim=1).to(x.device)
+
+        if self.refer_offset_t < 0:
+            self.refer_offset_t = grid_sizes[0][0].item()
+        if self.refer_offset_h < 0:
+            self.refer_offset_h = grid_sizes[0][1].item()
+        if self.refer_offset_w < 0:
+            self.refer_offset_w = grid_sizes[0][2].item()
+
+        self.freqs_ref = torch.cat([
+            rope_params(512, d - 4 * (d // 6), offset=self.refer_offset_t),
+            rope_params(512, 2 * (d // 6), offset=self.refer_offset_h),
+            rope_params(512, 2 * (d // 6), offset=self.refer_offset_w)
+        ], dim=1).to(x.device)
+
+        # time embeddings
+        e = self.time_embedding(sinusoidal_embedding_1d(self.freq_dim, t).to(x.dtype))
+        e0 = self.time_projection(e).unflatten(1, (6, self.dim))
+
+        # [context]
+        context_lens = None
+        context = self.text_embedding(torch.stack([torch.cat([
+            u, u.new_zeros(self.text_len - u.size(0), u.size(1))
+        ]) for u in context]))
+
+        if self.use_img_emb:
+            context_clip = self.img_emb(clip_fea) # bs x 257 x dim
+            context = torch.concat([context_clip, context], dim=1)
+
+        block_mask_id = (origin_len, origin_area[0], origin_area[1])
+        if block_mask_id not in self.block_masks:
+            self.block_masks[block_mask_id] = self.create_mask(origin_len, origin_area, x.device)
+        attn_mask = self.block_masks[block_mask_id]
+
+        # arguments
+        kwargs = dict(
+            e=e0,
+            attn_mask=attn_mask,
+            grid_sizes=grid_sizes,
+            freqs=self.freqs,
+            context=context,
+            grid_sizes_ref=grid_sizes_ref,
+            freqs_ref=self.freqs_ref,
+            context_lens=context_lens,
+            origin_area=origin_area,
+            origin_len=origin_len,
+            use_context_parallel=use_unified_sequence_parallel,
+            log_scale=log_scale,
+        )
+        if use_unified_sequence_parallel:
+            from ..utils.xfuser import get_current_chunk, is_evenly_divisible, get_sequence_parallel_world_size
+            assert is_evenly_divisible(x.shape[1]), (
+                f"sequence length ({x.shape[1]}) must be divisible by sequence parallel world size ({get_sequence_parallel_world_size()}) "
+            )
+            x = get_current_chunk(x, dim=1)
+
+        for idx, block in enumerate(self.blocks):
+            if is_uncondtion and idx==9:
+                continue
+            x = block(x, idx, k_cache, v_cache, method='forward_gen', **kwargs)
+
+        # head
+        x = self.head(x, e)
+        if use_unified_sequence_parallel:
+            from ..utils.xfuser import gather_all_chunks
+            x = gather_all_chunks(x, dim=1)
+
+        # unpatchify
+        x = self.unpatchify(x, grid_sizes)
+        return [u for u in x]
+
+    def forward_origin(
+        self,
+        x,
+        clip_fea,
+        y,
+        context,
+        seq_len,
+        x_ref,
+        clip_fea_ref,
+        y_ref,
+        context_ref,
+        seq_len_ref,
+        t,
+        origin_len,
+        origin_area,
+        log_scale=0.0,
+        use_gradient_checkpointing: bool = False,
+        use_gradient_checkpointing_offload: bool = False,
+    ):
+        # params
+        x = [torch.cat([u, v], dim=0) for u, v in zip(x, y)]
+        # embeddings
+        x = [self.patch_embedding(u.unsqueeze(0)) for u in x]
+        grid_sizes = torch.stack([
+            torch.tensor(u.shape[2:], dtype=torch.long) for u in x
+        ])
+        x = [u.flatten(2).transpose(1, 2) for u in x]
+        seq_lens = torch.tensor([u.size(1) for u in x], dtype=torch.long)
+        assert seq_lens.max() <= seq_len
+        x = torch.cat([torch.cat([
+            u, u.new_zeros(1, seq_len - u.size(1), u.size(2))
+        ], dim=1) for u in x])
+
+        # [reference]
+        # params
+        x_ref = [torch.cat([u, v], dim=0) for u, v in zip(x_ref, y_ref)]
+        # embeddings
+        x_ref = [self.patch_embedding(u.unsqueeze(0)) for u in x_ref]
+        grid_sizes_ref = torch.stack([
+            torch.tensor(u.shape[2:], dtype=torch.long) for u in x_ref
+        ])
+        x_ref = [u.flatten(2).transpose(1, 2) for u in x_ref]
+        seq_lens_ref = torch.tensor([u.size(1) for u in x_ref], dtype=torch.long)
+        assert seq_lens_ref.max() <= seq_len_ref
+        x_ref = torch.cat([torch.cat([
+            u, u.new_zeros(1, seq_len_ref - u.size(1), u.size(2))
+        ], dim=1) for u in x_ref])
+
+        assert (self.dim % self.num_heads) == 0 and (self.dim // self.num_heads) % 2 == 0
+        d = self.dim // self.num_heads
+        self.freqs = torch.cat([
+            rope_params(512, d - 4 * (d // 6)),
+            rope_params(512, 2 * (d // 6)),
+            rope_params(512, 2 * (d // 6))
+        ], dim=1).to(x.device)
+
+        if self.refer_offset_t < 0:
+            self.refer_offset_t = grid_sizes[0][0].item()
+        if self.refer_offset_h < 0:
+            self.refer_offset_h = grid_sizes[0][1].item()
+        if self.refer_offset_w < 0:
+            self.refer_offset_w = grid_sizes[0][2].item()
+
+        self.freqs_ref = torch.cat([
+            rope_params(512, d - 4 * (d // 6), offset=self.refer_offset_t),
+            rope_params(512, 2 * (d // 6), offset=self.refer_offset_h),
+            rope_params(512, 2 * (d // 6), offset=self.refer_offset_w)
+        ], dim=1).to(x.device)
+
+        # time embeddings
+        e = self.time_embedding(sinusoidal_embedding_1d(self.freq_dim, t).to(x.dtype))
+        e0 = self.time_projection(e).unflatten(1, (6, self.dim))
+
+        # time embeddings ref
+        e_ref = self.time_embedding(sinusoidal_embedding_1d(self.freq_dim, t*0+1).to(x.dtype))
+        e0_ref = self.time_projection(e_ref).unflatten(1, (6, self.dim))
+
+        # [context]
+        context_lens = None
+        context = self.text_embedding(torch.stack([torch.cat([
+            u, u.new_zeros(self.text_len - u.size(0), u.size(1))
+        ]) for u in context]))
+
+        if self.use_img_emb:
+            context_clip = self.img_emb(clip_fea) # bs x 257 x dim
+            context = torch.concat([context_clip, context], dim=1)
+
+        # [context_ref]
+        context_ref = self.text_embedding(torch.stack([torch.cat([
+            u, u.new_zeros(self.text_len - u.size(0), u.size(1))
+        ]) for u in context_ref]))
+
+        if self.use_img_emb:
+            context_clip_ref = self.img_emb(clip_fea_ref) # bs x 257 x dim
+            context_ref = torch.concat([context_clip_ref, context_ref], dim=1)
+
+        block_mask_id = (origin_len, origin_area[0], origin_area[1])
+        if block_mask_id not in self.block_masks:
+            self.block_masks[block_mask_id] = self.create_mask(origin_len, origin_area, x.device)
+        attn_mask = self.block_masks[block_mask_id]
+
+        # arguments
+        ref_args = dict(
+            e_ref=e0_ref,
+            grid_sizes_ref=grid_sizes_ref,
+            freqs_ref=self.freqs_ref,
+            context_ref=context_ref,
+            context_lens=context_lens,
+        )
+        gen_args = dict(
+            e=e0,
+            attn_mask=attn_mask,
+            grid_sizes=grid_sizes,
+            freqs=self.freqs,
+            context=context,
+            grid_sizes_ref=grid_sizes_ref,
+            freqs_ref=self.freqs_ref,
+            context_lens=context_lens,
+            origin_area=origin_area,
+            origin_len=origin_len,
+            log_scale=log_scale,
+        )
+        for idx, block in enumerate(self.blocks):
+            x, x_ref = gradient_checkpoint_forward(
+                block,
+                use_gradient_checkpointing,
+                use_gradient_checkpointing_offload,
+                x, x_ref, ref_args, gen_args, method='forward_origin'
+            )
+
+        # head
+        x = self.head(x, e)
+
+        # unpatchify
+        x = self.unpatchify(x, grid_sizes)
+        return [u.float() for u in x]
+
+    def unpatchify(self, x, grid_sizes):
+        c = self.out_dim
+        out = []
+        for u, v in zip(x, grid_sizes.tolist()):
+            u = u[:math.prod(v)].view(*v, *self.patch_size, c)
+            u = torch.einsum('fhwpqrc->cfphqwr', u)
+            u = u.reshape(c, *[i * j for i, j in zip(v, self.patch_size)])
+            out.append(u)
+        return out
diff --git a/diffsynth/pipelines/wan_video.py b/diffsynth/pipelines/wan_video.py
index c1e4dfb3a..e2ab76f9c 100644
--- a/diffsynth/pipelines/wan_video.py
+++ b/diffsynth/pipelines/wan_video.py
@@ -1,4 +1,4 @@
-import torch, types
+import torch, types, math
 import numpy as np
 from PIL import Image
 from einops import repeat
@@ -27,6 +27,7 @@
 from ..models.wan_video_mot import MotWanModel
 from ..models.wav2vec import WanS2VAudioEncoder
 from ..models.longcat_video_dit import LongCatVideoTransformer3DModel
+from ..models.wan_animate_2_dit import WanAnimate2Transformer
 
 
 class WanVideoPipeline(BasePipeline):
@@ -78,9 +79,14 @@ def __init__(self, device=get_device_type(), torch_dtype=torch.bfloat16):
             WanVideoUnit_WanToDance_ProcessInputs(),
             WanVideoUnit_WanToDance_RefImageEmbedder(),
             WanVideoUnit_WanToDance_ImageKeyframesEmbedder(),
+            WanVideoUnit_Animate2RefPromptEmbedder(),
+            WanVideoUnit_Animate2CLIPEmbedder(),
+            WanVideoUnit_Animate2VAEEmbedder(),
+            WanVideoUnit_Animate2RefKVCacheEmbedder(),
         ]
         self.post_units = [
             WanVideoPostUnit_S2V(),
+            WanVideoPostUnit_Animate2(),
         ]
         self.model_fn = model_fn_wan_video
         self.compilable_models = ["dit", "dit2"]
@@ -88,6 +94,10 @@ def __init__(self, device=get_device_type(), torch_dtype=torch.bfloat16):
 
     def enable_usp(self):
         from ..utils.xfuser import get_sequence_parallel_world_size, usp_attn_forward, usp_dit_forward, usp_vace_forward
+        self.sp_size = get_sequence_parallel_world_size()
+        self.use_unified_sequence_parallel = True
+        if isinstance(self.dit, WanAnimate2Transformer):
+            return
 
         for block in self.dit.blocks:
             block.self_attn.forward = types.MethodType(usp_attn_forward, block.self_attn)
@@ -104,8 +114,6 @@ def enable_usp(self):
             for block in self.vace2.vace_blocks:
                 block.self_attn.forward = types.MethodType(usp_attn_forward, block.self_attn)
             self.vace2.forward = types.MethodType(usp_vace_forward, self.vace2)
-        self.sp_size = get_sequence_parallel_world_size()
-        self.use_unified_sequence_parallel = True
 
 
     @staticmethod
@@ -223,6 +231,13 @@ def __call__(
         animate_face_video: list[Image.Image] = None,
         animate_inpaint_video: list[Image.Image] = None,
         animate_mask_video: list[Image.Image] = None,
+        # Wan-Animate-2
+        animate2_prompt_ref: str = " ",
+        animate2_reference_image: Image.Image = None,
+        animate2_reference_video: list[Image.Image] = None,
+        animate2_refert_images: list[Image.Image] = None,
+        animate2_offload_kv: bool = False,
+        animate2_log_scale: float = 0.0,
         # VAP
         vap_video: list[Image.Image] = None,
         vap_prompt: str = " ",
@@ -273,11 +288,13 @@ def __call__(
         # Inputs
         inputs_posi = {
             "prompt": prompt,
+            "positive": True,
             "vap_prompt": vap_prompt,
             "tea_cache_l1_thresh": tea_cache_l1_thresh, "tea_cache_model_id": tea_cache_model_id, "num_inference_steps": num_inference_steps,
         }
         inputs_nega = {
             "negative_prompt": negative_prompt,
+            "positive": False,
             "negative_vap_prompt": negative_vap_prompt,
             "tea_cache_l1_thresh": tea_cache_l1_thresh, "tea_cache_model_id": tea_cache_model_id, "num_inference_steps": num_inference_steps,
         }
@@ -298,6 +315,7 @@ def __call__(
             "sliding_window_size": sliding_window_size, "sliding_window_stride": sliding_window_stride,
             "input_audio": input_audio, "audio_sample_rate": audio_sample_rate, "s2v_pose_video": s2v_pose_video, "audio_embeds": audio_embeds, "s2v_pose_latents": s2v_pose_latents, "motion_video": motion_video,
             "animate_pose_video": animate_pose_video, "animate_face_video": animate_face_video, "animate_inpaint_video": animate_inpaint_video, "animate_mask_video": animate_mask_video,
+            "animate2_prompt_ref": animate2_prompt_ref, "animate2_reference_image": animate2_reference_image, "animate2_reference_video": animate2_reference_video, "animate2_refert_images": animate2_refert_images, "animate2_offload_kv": animate2_offload_kv, "animate2_log_scale": animate2_log_scale,
             "vap_video": vap_video, 
             "wantodance_music_path": wantodance_music_path, "wantodance_reference_image": wantodance_reference_image, "wantodance_fps": wantodance_fps,
             "wantodance_keyframes": wantodance_keyframes, "wantodance_keyframes_mask": wantodance_keyframes_mask,
@@ -376,12 +394,14 @@ def process(self, pipe: WanVideoPipeline, height, width, num_frames):
 class WanVideoUnit_NoiseInitializer(PipelineUnit):
     def __init__(self):
         super().__init__(
-            input_params=("height", "width", "num_frames", "seed", "rand_device", "vace_reference_image"),
+            input_params=("height", "width", "num_frames", "seed", "rand_device", "vace_reference_image", "animate2_reference_video"),
             output_params=("noise",)
         )
 
-    def process(self, pipe: WanVideoPipeline, height, width, num_frames, seed, rand_device, vace_reference_image):
+    def process(self, pipe: WanVideoPipeline, height, width, num_frames, seed, rand_device, vace_reference_image, animate2_reference_video):
         length = (num_frames - 1) // 4 + 1
+        if animate2_reference_video is not None:
+            length += 1
         if vace_reference_image is not None:
             f = len(vace_reference_image) if isinstance(vace_reference_image, list) else 1
             length += f
@@ -390,18 +410,17 @@ def process(self, pipe: WanVideoPipeline, height, width, num_frames, seed, rand_
         if vace_reference_image is not None:
             noise = torch.concat((noise[:, :, -f:], noise[:, :, :-f]), dim=2)
         return {"noise": noise}
-    
 
 
 class WanVideoUnit_InputVideoEmbedder(PipelineUnit):
     def __init__(self):
         super().__init__(
-            input_params=("input_video", "noise", "tiled", "tile_size", "tile_stride", "vace_reference_image", "framewise_decoding"),
+            input_params=("input_video", "noise", "tiled", "tile_size", "tile_stride", "vace_reference_image", "framewise_decoding", "animate2_reference_image", "animate2_reference_video"),
             output_params=("latents", "input_latents"),
             onload_model_names=("vae",)
         )
 
-    def process(self, pipe: WanVideoPipeline, input_video, noise, tiled, tile_size, tile_stride, vace_reference_image, framewise_decoding):
+    def process(self, pipe: WanVideoPipeline, input_video, noise, tiled, tile_size, tile_stride, vace_reference_image, framewise_decoding, animate2_reference_image, animate2_reference_video):
         if input_video is None:
             return {"latents": noise}
         pipe.load_models_to_device(self.onload_model_names)
@@ -417,6 +436,12 @@ def process(self, pipe: WanVideoPipeline, input_video, noise, tiled, tile_size,
             vace_reference_latents = pipe.vae.encode(vace_reference_image, device=pipe.device).to(dtype=pipe.torch_dtype, device=pipe.device)
             input_latents = torch.concat([vace_reference_latents, input_latents], dim=2)
         if pipe.scheduler.training:
+            if animate2_reference_image is not None and animate2_reference_video is not None:
+                vh, vw = input_video.shape[-2], input_video.shape[-1]
+                ref_img = pipe.preprocess_image(animate2_reference_image.resize((vw, vh))).to(pipe.device)  # (1, C, H, W)
+                ref_pixel = ref_img.transpose(0, 1)  # (C, 1, H, W)
+                ref_latent = pipe.vae.encode([ref_pixel.to(pipe.torch_dtype)], device=pipe.device).to(dtype=pipe.torch_dtype, device=pipe.device)
+                input_latents = torch.concat([ref_latent, input_latents], dim=2)
             return {"latents": noise, "input_latents": input_latents}
         else:
             latents = pipe.scheduler.add_noise(input_latents, noise, timestep=pipe.scheduler.timesteps[0])
@@ -1151,6 +1176,160 @@ def process(self, pipe: WanVideoPipeline, wantodance_keyframes, wantodance_keyfr
         return {"clip_feature": clip_context, "y": y}
 
 
+class WanVideoUnit_Animate2RefPromptEmbedder(PipelineUnit):
+    def __init__(self):
+        super().__init__(
+            input_params=("animate2_prompt_ref", "animate2_reference_video"),
+            output_params=("context_ref",),
+            onload_model_names=("text_encoder",)
+        )
+
+    def process(self, pipe: WanVideoPipeline, animate2_prompt_ref, animate2_reference_video):
+        if animate2_reference_video is None or animate2_prompt_ref is None:
+            return {}
+        pipe.load_models_to_device(self.onload_model_names)
+        context_ref = WanVideoUnit_PromptEmbedder().encode_prompt(pipe, animate2_prompt_ref)
+        return {"context_ref": context_ref}
+
+
+class WanVideoUnit_Animate2CLIPEmbedder(PipelineUnit):
+    def __init__(self):
+        super().__init__(
+            input_params=("animate2_reference_image", "animate2_reference_video", "height", "width"),
+            output_params=("clip_feature", "clip_fea_ref"),
+            onload_model_names=("image_encoder",)
+        )
+
+    def process(self, pipe: WanVideoPipeline, animate2_reference_image, animate2_reference_video, height, width):
+        if animate2_reference_image is None or animate2_reference_video is None:
+            return {}
+        pipe.load_models_to_device(self.onload_model_names)
+        ref = pipe.preprocess_image(animate2_reference_image.resize((width, height))).to(pipe.device)
+        clip_feature = pipe.image_encoder.encode_image([ref]).to(dtype=pipe.torch_dtype, device=pipe.device)
+        ref_video_0 = pipe.preprocess_image(animate2_reference_video[0].resize((width, height))).to(pipe.device)
+        clip_fea_ref = pipe.image_encoder.encode_image([ref_video_0]).to(dtype=pipe.torch_dtype, device=pipe.device)
+        return {"clip_feature": clip_feature, "clip_fea_ref": clip_fea_ref}
+
+
+class WanVideoUnit_Animate2VAEEmbedder(PipelineUnit):
+    def __init__(self):
+        super().__init__(
+            input_params=("animate2_reference_image", "animate2_reference_video", "animate2_refert_images", "num_frames", "height", "width", "tiled", "tile_size", "tile_stride"),
+            output_params=("y", "condition_latents", "condition_y", "grid_sizes", "grid_sizes_ref", "seq_len", "seq_len_ref", "origin_len", "origin_area"),
+            onload_model_names=("vae",)
+        )
+
+    @staticmethod
+    def animate2_get_i2v_mask(lat_t, lat_h, lat_w, mask_len=1, device="cuda"):
+        msk = torch.zeros(1, (lat_t - 1) * 4 + 1, lat_h, lat_w, device=device)
+        msk[:, :mask_len] = 1
+        msk = torch.concat([torch.repeat_interleave(msk[:, 0:1], repeats=4, dim=1), msk[:, 1:]], dim=1)
+        msk = msk.view(1, msk.shape[1] // 4, 4, lat_h, lat_w)
+        msk = msk.transpose(1, 2)[0]
+        return msk
+
+    def process(self, pipe: WanVideoPipeline, animate2_reference_image, animate2_reference_video, animate2_refert_images, num_frames, height, width, tiled, tile_size, tile_stride):
+        if animate2_reference_image is None or animate2_reference_video is None:
+            return {}
+        pipe.load_models_to_device(self.onload_model_names)
+        device, dtype = pipe.device, pipe.torch_dtype
+        H, W = height, width
+        T = num_frames + 1
+        lat_h, lat_w = H // 8, W // 8
+        lat_t = T // 4 + 1 + 1
+
+        # Reference image -> y_ref (mask + latent), single latent frame
+        ref_img = pipe.preprocess_image(animate2_reference_image.resize((W, H))).to(device)  # (1, C, H, W)
+        ref_pixel = ref_img.transpose(0, 1)  # (C, 1, H, W)
+        ref_latents = pipe.vae.encode([ref_pixel.to(dtype)], device, tiled, tile_size, tile_stride).to(dtype=dtype, device=device)  # (1, 16, 1, lat_h, lat_w)
+        mask_ref = self.animate2_get_i2v_mask(1, lat_h, lat_w, mask_len=1, device=device)
+        y_ref = torch.concat([mask_ref, ref_latents[0]]).to(dtype=dtype, device=device)  # (20, 1, lat_h, lat_w)
+
+        # Reference-temporal slot: for multi-clip long video, the first `mask_reft_len` temporal
+        # positions carry the previous clip's tail frames (continuation); the rest are zeros.
+        # Single/first clip: animate2_refert_images is None -> mask_reft_len == 0 (all zeros).
+        if animate2_refert_images is not None and len(animate2_refert_images) > 0:
+            mask_reft_len = len(animate2_refert_images)
+            refert = pipe.preprocess_video([f.resize((W, H)) for f in animate2_refert_images]).to(device)  # (1, 3, mask_reft_len, H, W)
+            zeros_tail = torch.zeros(3, T - 1 - mask_reft_len, H, W, device=device, dtype=dtype)
+            reft_vid = torch.concat([refert[0].to(dtype), zeros_tail], dim=1)  # (3, T-1, H, W)
+        else:
+            mask_reft_len = 0
+            reft_vid = torch.zeros(3, T - 1, H, W, device=device, dtype=dtype)
+        y_reft = pipe.vae.encode([reft_vid], device, tiled, tile_size, tile_stride)[0].to(dtype=dtype, device=device)  # (16, lat_t-1, lat_h, lat_w)
+        msk_reft = self.animate2_get_i2v_mask(lat_t - 1, lat_h, lat_w, mask_len=mask_reft_len, device=device)
+        y_reft = torch.concat([msk_reft, y_reft]).to(dtype=dtype, device=device)
+        y = torch.concat([y_ref, y_reft], dim=1)  # (20, lat_t, lat_h, lat_w)
+
+        # Reference video -> condition_latents
+        ref_video = pipe.preprocess_video([f.resize((W, H)) for f in animate2_reference_video[:num_frames]]).to(device)  # (1, C, num_frames, H, W)
+        condition_latents = pipe.vae.encode([ref_video[0].to(dtype)], device, tiled, tile_size, tile_stride).to(dtype=dtype, device=device)  # (1, 16, lat_t_c, lat_h, lat_w)
+        # Reference video -> condition_y (mask + latent), mask_len = T
+        _, _, lat_t_c, lat_h_c, lat_w_c = condition_latents.shape
+        condition_y = condition_latents.clone()[0]
+        condition_msk_y = self.animate2_get_i2v_mask(lat_t_c, lat_h_c, lat_w_c, mask_len=ref_video.shape[2], device=device)
+        condition_y = torch.concat([condition_msk_y, condition_y]).to(dtype=dtype, device=device)  # (20, lat_t_c, lat_h_c, lat_w_c)
+
+        # sequence lengths
+        grid_sizes = torch.stack([torch.tensor([lat_t, lat_h // 2, lat_w // 2], dtype=torch.long)])
+        grid_sizes_ref = torch.stack([torch.tensor([lat_t_c, lat_h_c // 2, lat_w_c // 2], dtype=torch.long)])
+        seq_len = int(math.ceil(lat_t * lat_h * lat_w / 4))
+        seq_len_ref = int(math.ceil(lat_t_c * lat_h * lat_w / 4))
+
+        return {
+            "y": y.unsqueeze(0),
+            "condition_latents": condition_latents,
+            "condition_y": condition_y,
+            "grid_sizes": grid_sizes,
+            "grid_sizes_ref": grid_sizes_ref,
+            "seq_len": seq_len,
+            "seq_len_ref": seq_len_ref,
+            "origin_len": num_frames,
+            "origin_area": [W, H],
+        }
+
+
+class WanVideoUnit_Animate2RefKVCacheEmbedder(PipelineUnit):
+    def __init__(self):
+        super().__init__(
+            input_params=("animate2_reference_video", "condition_latents", "condition_y", "context_ref", "clip_fea_ref", "grid_sizes", "seq_len_ref", "animate2_offload_kv", "use_gradient_checkpointing", "use_gradient_checkpointing_offload", "use_unified_sequence_parallel"),
+            output_params=("animate2_k_cache", "animate2_v_cache"),
+            onload_model_names=("dit",)
+        )
+
+    def process(self, pipe: WanVideoPipeline, animate2_reference_video, condition_latents, condition_y, context_ref, clip_fea_ref, grid_sizes, seq_len_ref, animate2_offload_kv, use_gradient_checkpointing, use_gradient_checkpointing_offload, use_unified_sequence_parallel):
+        if animate2_reference_video is None or pipe.scheduler.training:
+            return {}
+        pipe.load_models_to_device(self.onload_model_names)
+        animate2_k_cache, animate2_v_cache = {}, {}
+        t = pipe.scheduler.timesteps[0]
+        timestep = t.unsqueeze(0).to(dtype=pipe.torch_dtype, device=pipe.device)
+        pipe.dit.forward_ref(
+            condition_latents,
+            grid_sizes=grid_sizes.to(pipe.device),
+            k_cache=animate2_k_cache,
+            v_cache=animate2_v_cache,
+            clip_fea_ref=clip_fea_ref,
+            y_ref=[condition_y],
+            context_ref=[context_ref[0]],
+            seq_len_ref=seq_len_ref,
+            t=timestep,
+            animate2_offload_kv=animate2_offload_kv,
+            use_unified_sequence_parallel=use_unified_sequence_parallel,
+        )
+        return {"animate2_k_cache": animate2_k_cache, "animate2_v_cache": animate2_v_cache}
+
+
+class WanVideoPostUnit_Animate2(PipelineUnit):
+    def __init__(self):
+        super().__init__(input_params=("latents", "animate2_reference_video"))
+
+    def process(self, pipe: WanVideoPipeline, latents, animate2_reference_video):
+        if animate2_reference_video is None:
+            return {}
+        return {"latents": latents[:, :, 1:].to(pipe.torch_dtype)}
+
+
 class TeaCache:
     def __init__(self, num_inference_steps, rel_l1_thresh, model_id):
         self.num_inference_steps = num_inference_steps
@@ -1349,6 +1528,15 @@ def model_fn_wan_video(
             use_gradient_checkpointing=use_gradient_checkpointing,
             use_gradient_checkpointing_offload=use_gradient_checkpointing_offload,
         )
+    # Wan-Animate-2
+    if isinstance(dit, WanAnimate2Transformer):
+        return model_fn_wananimate(
+            dit=dit, latents=latents, timestep=timestep, context=context,
+            clip_feature=clip_feature, y=y, use_unified_sequence_parallel=use_unified_sequence_parallel,
+            use_gradient_checkpointing=use_gradient_checkpointing,
+            use_gradient_checkpointing_offload=use_gradient_checkpointing_offload,
+            **kwargs,
+        )
         
     # wan2.2 s2v
     if audio_embeds is not None:
@@ -1715,3 +1903,70 @@ def custom_forward(*inputs):
     # make compatible with wan video
     x = torch.cat([origin_ref_latents, x], dim=2)
     return x
+
+
+def model_fn_wananimate(
+    dit: WanAnimate2Transformer,
+    latents: torch.Tensor = None,
+    timestep: torch.Tensor = None,
+    context: torch.Tensor = None,
+    clip_feature: torch.Tensor = None,
+    y: torch.Tensor = None,
+    animate2_k_cache: dict = None,
+    animate2_v_cache: dict = None,
+    grid_sizes_ref: torch.Tensor = None,
+    origin_len: int = None,
+    origin_area: list = None,
+    seq_len: int = None,
+    positive: bool = True,
+    condition_latents: torch.Tensor = None,
+    clip_fea_ref: torch.Tensor = None,
+    condition_y: torch.Tensor = None,
+    context_ref: torch.Tensor = None,
+    seq_len_ref: int = None,
+    use_unified_sequence_parallel: bool = False,
+    animate2_log_scale: float = 0.0,
+    use_gradient_checkpointing_offload: bool = False,
+    use_gradient_checkpointing: bool = False,
+    **kwargs,
+):
+    is_uncondtion = not positive
+    if animate2_k_cache is None or animate2_v_cache is None:
+        # training
+        out = dit.forward_origin(
+            x=[latents[0]],
+            clip_fea=clip_feature,
+            y=[y[0]],
+            context=[context[0]],
+            seq_len=seq_len,
+            x_ref=condition_latents,
+            clip_fea_ref=clip_fea_ref,
+            y_ref=[condition_y],
+            context_ref=[context_ref[0]],
+            seq_len_ref=seq_len_ref,
+            t=timestep,
+            origin_len=origin_len,
+            origin_area=origin_area,
+            log_scale=animate2_log_scale,
+            use_gradient_checkpointing=use_gradient_checkpointing,
+            use_gradient_checkpointing_offload=use_gradient_checkpointing_offload,
+        )
+    else:
+        # inference
+        out = dit.forward_gen(
+            x=[latents[0]],
+            k_cache=animate2_k_cache,
+            v_cache=animate2_v_cache,
+            clip_fea=clip_feature,
+            y=[y[0]],
+            context=[context[0]],
+            seq_len=seq_len,
+            t=timestep,
+            grid_sizes_ref=grid_sizes_ref.to(latents.device),
+            origin_len=origin_len,
+            origin_area=origin_area,
+            is_uncondtion=is_uncondtion,
+            use_unified_sequence_parallel=use_unified_sequence_parallel,
+            log_scale=animate2_log_scale,
+        )
+    return out[0].unsqueeze(0)
diff --git a/diffsynth/utils/xfuser/__init__.py b/diffsynth/utils/xfuser/__init__.py
index cf6025b7e..8a3a4f1f2 100644
--- a/diffsynth/utils/xfuser/__init__.py
+++ b/diffsynth/utils/xfuser/__init__.py
@@ -1 +1 @@
-from .xdit_context_parallel import usp_attn_forward, usp_dit_forward, usp_vace_forward, get_sequence_parallel_world_size, get_sequence_parallel_rank, get_sp_group, initialize_usp, get_current_chunk, gather_all_chunks
+from .xdit_context_parallel import usp_attn_forward, usp_dit_forward, usp_vace_forward, get_sequence_parallel_world_size, get_sequence_parallel_rank, get_sp_group, initialize_usp, get_current_chunk, gather_all_chunks, all_to_all_4d, is_evenly_divisible
diff --git a/diffsynth/utils/xfuser/xdit_context_parallel.py b/diffsynth/utils/xfuser/xdit_context_parallel.py
index abf0f3fef..23b8508b9 100644
--- a/diffsynth/utils/xfuser/xdit_context_parallel.py
+++ b/diffsynth/utils/xfuser/xdit_context_parallel.py
@@ -2,6 +2,7 @@
 from typing import Optional
 from einops import rearrange
 from yunchang.kernels import AttnType
+from yunchang.comm.all_to_all import SeqAllToAll4D
 from xfuser.core.distributed import (get_sequence_parallel_rank,
                                      get_sequence_parallel_world_size,
                                      get_sp_group)
@@ -204,3 +205,15 @@ def gather_all_chunks(x, seq_len=None, dim=1):
         slices[dim] = slice(0, seq_len)
         x = x[tuple(slices)]
     return x
+
+
+def all_to_all_4d(x, scatter_dim, gather_dim):
+    world_size = get_sequence_parallel_world_size()
+    if world_size == 1:
+        return x
+    return SeqAllToAll4D.apply(get_sp_group().ulysses_group, x, scatter_dim, gather_dim)
+
+
+def is_evenly_divisible(seq_len):
+    world_size = get_sequence_parallel_world_size()
+    return seq_len % world_size == 0
diff --git a/docs/en/Model_Details/Wan.md b/docs/en/Model_Details/Wan.md
index ce5ca3a3c..632f007b8 100644
--- a/docs/en/Model_Details/Wan.md
+++ b/docs/en/Model_Details/Wan.md
@@ -91,6 +91,7 @@ graph LR;
     Wan-AI/Wan2.1-T2V-14B-->krea/krea-realtime-video;
     Wan-AI/Wan2.1-I2V-14B-720P-->ByteDance/Video-As-Prompt-Wan2.1-14B;
     Wan-AI/Wan2.1-T2V-14B-->Wan-AI/Wan2.2-Animate-14B;
+    Wan2.2-Series-->Wan-AI/Wan2.2-Animate-2-14B;
     Wan-AI/Wan2.1-T2V-14B-->Wan-AI/Wan2.2-S2V-14B;
     Wan2.2-Series-->Wan-AI/Wan2.2-T2V-A14B;
     Wan2.2-Series-->Wan-AI/Wan2.2-I2V-A14B;
@@ -132,6 +133,8 @@ graph LR;
 |[Wan-AI/Wan2.2-I2V-A14B](https://modelscope.cn/models/Wan-AI/Wan2.2-I2V-A14B)|`input_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.2-I2V-A14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.2-I2V-A14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.2-I2V-A14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.2-I2V-A14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.2-I2V-A14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.2-I2V-A14B.py)|
 |[Wan-AI/Wan2.2-TI2V-5B](https://modelscope.cn/models/Wan-AI/Wan2.2-TI2V-5B)|`input_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.2-TI2V-5B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.2-TI2V-5B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.2-TI2V-5B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.2-TI2V-5B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.2-TI2V-5B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.2-TI2V-5B.py)|
 |[Wan-AI/Wan2.2-Animate-14B](https://www.modelscope.cn/models/Wan-AI/Wan2.2-Animate-14B)|`input_image`, `animate_pose_video`, `animate_face_video`, `animate_inpaint_video`, `animate_mask_video`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.2-Animate-14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.2-Animate-14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.2-Animate-14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.2-Animate-14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.2-Animate-14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.2-Animate-14B.py)|
+|[Wan-AI/Wan2.2-Animate-2-14B](https://www.modelscope.cn/models/Wan-AI/Wan2.2-Animate-2-14B)|`animate2_reference_image`, `animate2_reference_video`, `animate2_prompt_ref`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.2-Animate-2-14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.2-Animate-2-14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.2-Animate-2-14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.2-Animate-2-14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.2-Animate-2-14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.2-Animate-2-14B.py)|
+|[Wan-AI/Wan2.2-Animate-2-14B: Distilled](https://www.modelscope.cn/models/Wan-AI/Wan2.2-Animate-2-14B)|`animate2_reference_image`, `animate2_reference_video`, `animate2_prompt_ref`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.2-Animate-2-14B-Distilled.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.2-Animate-2-14B-Distilled.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.2-Animate-2-14B-Distilled.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.2-Animate-2-14B-Distilled.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.2-Animate-2-14B-Distilled.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.2-Animate-2-14B-Distilled.py)|
 |[Wan-AI/Wan2.2-S2V-14B](https://www.modelscope.cn/models/Wan-AI/Wan2.2-S2V-14B)|`input_image`, `input_audio`, `audio_sample_rate`, `s2v_pose_video`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.2-S2V-14B_multi_clips.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.2-S2V-14B_multi_clips.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.2-S2V-14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.2-S2V-14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.2-S2V-14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.2-S2V-14B.py)|
 |[PAI/Wan2.2-VACE-Fun-A14B](https://www.modelscope.cn/models/PAI/Wan2.2-VACE-Fun-A14B)|`vace_control_video`, `vace_reference_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.2-VACE-Fun-A14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.2-VACE-Fun-A14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.2-VACE-Fun-A14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.2-VACE-Fun-A14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.2-VACE-Fun-A14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.2-VACE-Fun-A14B.py)|
 |[PAI/Wan2.2-Fun-A14B-InP](https://modelscope.cn/models/PAI/Wan2.2-Fun-A14B-InP)|`input_image`, `end_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.2-Fun-A14B-InP.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.2-Fun-A14B-InP.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.2-Fun-A14B-InP.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.2-Fun-A14B-InP.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.2-Fun-A14B-InP.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.2-Fun-A14B-InP.py)|
@@ -184,6 +187,12 @@ Input parameters for `WanVideoPipeline` inference include:
 * `audio_sample_rate`: Audio sampling rate, default value is 16000.
 * `s2v_pose_video`: S2V model pose video.
 * `motion_video`: S2V model motion video.
+* `animate2_reference_image`: Wan-Animate-2 reference image, providing the character identity.
+* `animate2_reference_video`: Wan-Animate-2 driving video, providing the motion.
+* `animate2_prompt_ref`: Wan-Animate-2 reference prompt for the driving video, describing its content.
+* `animate2_refert_images`: Wan-Animate-2 reference frame images for temporal continuation, used in long video chunked generation.
+* `animate2_offload_kv`: Whether Wan-Animate-2 offloads the reference video KV cache to memory, default value is `False`.
+* `animate2_log_scale`: Wan-Animate-2 guidance log scale, default value is 0.0, recommended to set to -1.3 for the distilled model.
 * `height`: Video height, must be a multiple of 16.
 * `width`: Video width, must be a multiple of 16.
 * `num_frames`: Number of video frames, default value is 81, must be a multiple of 4 + 1.
diff --git a/docs/zh/Model_Details/Wan.md b/docs/zh/Model_Details/Wan.md
index afda86162..55fd0885f 100644
--- a/docs/zh/Model_Details/Wan.md
+++ b/docs/zh/Model_Details/Wan.md
@@ -92,6 +92,7 @@ graph LR;
     Wan-AI/Wan2.1-T2V-14B-->meituan-longcat/LongCat-Video;
     Wan-AI/Wan2.1-I2V-14B-720P-->ByteDance/Video-As-Prompt-Wan2.1-14B;
     Wan-AI/Wan2.1-T2V-14B-->Wan-AI/Wan2.2-Animate-14B;
+    Wan2.2-Series-->Wan-AI/Wan2.2-Animate-2-14B;
     Wan-AI/Wan2.1-T2V-14B-->Wan-AI/Wan2.2-S2V-14B;
     Wan2.2-Series-->Wan-AI/Wan2.2-T2V-A14B;
     Wan2.2-Series-->Wan-AI/Wan2.2-I2V-A14B;
@@ -133,6 +134,8 @@ graph LR;
 |[Wan-AI/Wan2.2-I2V-A14B](https://modelscope.cn/models/Wan-AI/Wan2.2-I2V-A14B)|`input_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.2-I2V-A14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.2-I2V-A14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.2-I2V-A14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.2-I2V-A14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.2-I2V-A14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.2-I2V-A14B.py)|
 |[Wan-AI/Wan2.2-TI2V-5B](https://modelscope.cn/models/Wan-AI/Wan2.2-TI2V-5B)|`input_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.2-TI2V-5B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.2-TI2V-5B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.2-TI2V-5B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.2-TI2V-5B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.2-TI2V-5B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.2-TI2V-5B.py)|
 |[Wan-AI/Wan2.2-Animate-14B](https://www.modelscope.cn/models/Wan-AI/Wan2.2-Animate-14B)|`input_image`, `animate_pose_video`, `animate_face_video`, `animate_inpaint_video`, `animate_mask_video`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.2-Animate-14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.2-Animate-14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.2-Animate-14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.2-Animate-14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.2-Animate-14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.2-Animate-14B.py)|
+|[Wan-AI/Wan2.2-Animate-2-14B](https://www.modelscope.cn/models/Wan-AI/Wan2.2-Animate-2-14B)|`animate2_reference_image`, `animate2_reference_video`, `animate2_prompt_ref`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.2-Animate-2-14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.2-Animate-2-14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.2-Animate-2-14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.2-Animate-2-14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.2-Animate-2-14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.2-Animate-2-14B.py)|
+|[Wan-AI/Wan2.2-Animate-2-14B: Distilled](https://www.modelscope.cn/models/Wan-AI/Wan2.2-Animate-2-14B)|`animate2_reference_image`, `animate2_reference_video`, `animate2_prompt_ref`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.2-Animate-2-14B-Distilled.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.2-Animate-2-14B-Distilled.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.2-Animate-2-14B-Distilled.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.2-Animate-2-14B-Distilled.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.2-Animate-2-14B-Distilled.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.2-Animate-2-14B-Distilled.py)|
 |[Wan-AI/Wan2.2-S2V-14B](https://www.modelscope.cn/models/Wan-AI/Wan2.2-S2V-14B)|`input_image`, `input_audio`, `audio_sample_rate`, `s2v_pose_video`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.2-S2V-14B_multi_clips.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.2-S2V-14B_multi_clips.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.2-S2V-14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.2-S2V-14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.2-S2V-14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.2-S2V-14B.py)|
 |[PAI/Wan2.2-VACE-Fun-A14B](https://www.modelscope.cn/models/PAI/Wan2.2-VACE-Fun-A14B)|`vace_control_video`, `vace_reference_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.2-VACE-Fun-A14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.2-VACE-Fun-A14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.2-VACE-Fun-A14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.2-VACE-Fun-A14B.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.2-VACE-Fun-A14B.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.2-VACE-Fun-A14B.py)|
 |[PAI/Wan2.2-Fun-A14B-InP](https://modelscope.cn/models/PAI/Wan2.2-Fun-A14B-InP)|`input_image`, `end_image`|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference/Wan2.2-Fun-A14B-InP.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_inference_low_vram/Wan2.2-Fun-A14B-InP.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/full/Wan2.2-Fun-A14B-InP.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_full/Wan2.2-Fun-A14B-InP.py)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/lora/Wan2.2-Fun-A14B-InP.sh)|[code](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/wanvideo/model_training/validate_lora/Wan2.2-Fun-A14B-InP.py)|
@@ -185,6 +188,12 @@ DeepSpeed ZeRO 3 训练：Wan 系列模型支持 DeepSpeed ZeRO 3 训练，将
 * `audio_sample_rate`: 音频采样率，默认值为 16000。
 * `s2v_pose_video`: S2V 模型的姿态视频。
 * `motion_video`: S2V 模型的运动视频。
+* `animate2_reference_image`: Wan-Animate-2 模型的参考图像，提供角色身份。
+* `animate2_reference_video`: Wan-Animate-2 模型的驱动视频，提供动作。
+* `animate2_prompt_ref`: Wan-Animate-2 模型驱动视频的参考提示词，描述驱动视频中的内容。
+* `animate2_refert_images`: Wan-Animate-2 模型用于时序续接的参考帧图像，用于长视频分块生成。
+* `animate2_offload_kv`: Wan-Animate-2 模型是否将参考视频的 KV 缓存卸载到内存，默认值为 `False`。
+* `animate2_log_scale`: Wan-Animate-2 模型的引导对数缩放系数，默认值为 0.0，蒸馏模型建议设为 -1.3。
 * `height`: 视频高度，需保证高度为 16 的倍数。
 * `width`: 视频宽度，需保证宽度为 16 的倍数。
 * `num_frames`: 视频帧数，默认值为 81，需保证为 4 的倍数 + 1。
diff --git a/examples/wanvideo/acceleration/Wan2.2-Animate-2-14B-usp.py b/examples/wanvideo/acceleration/Wan2.2-Animate-2-14B-usp.py
new file mode 100644
index 000000000..8553a8a1b
--- /dev/null
+++ b/examples/wanvideo/acceleration/Wan2.2-Animate-2-14B-usp.py
@@ -0,0 +1,117 @@
+import torch
+import torch.distributed as dist
+from PIL import Image
+from diffsynth.utils.data import save_video, VideoData
+from diffsynth.pipelines.wan_video import WanVideoPipeline, ModelConfig
+from modelscope import dataset_snapshot_download
+
+vram_config = {
+    "offload_dtype": torch.bfloat16,
+    "offload_device": "cpu",
+    "onload_dtype": torch.bfloat16,
+    "onload_device": "cuda",
+    "preparing_dtype": torch.bfloat16,
+    "preparing_device": "cuda",
+    "computation_dtype": torch.bfloat16,
+    "computation_device": "cuda",
+}
+
+pipe = WanVideoPipeline.from_pretrained(
+    torch_dtype=torch.bfloat16,
+    device="cuda",
+    use_usp=True,
+    model_configs=[
+        ModelConfig(model_id="Wan-AI/Wan2.2-Animate-2-14B", origin_file_pattern="wan_animate_2/wan_animate_2_bf16.safetensors", **vram_config),
+        ModelConfig(model_id="Wan-AI/Wan2.2-Animate-2-14B", origin_file_pattern="videomodel/Wan-AI/models_t5_umt5-xxl-enc-bf16.pth", **vram_config),
+        ModelConfig(model_id="Wan-AI/Wan2.2-Animate-2-14B", origin_file_pattern="videomodel/Wan-AI/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth", **vram_config),
+        ModelConfig(model_id="Wan-AI/Wan2.1-T2V-14B", origin_file_pattern="Wan2.1_VAE.pth", **vram_config),
+    ],
+    tokenizer_config=ModelConfig(model_id="Wan-AI/Wan2.2-Animate-2-14B", origin_file_pattern="videomodel/Wan-AI/umt5-xxl/"),
+)
+
+# Character animation: reference image (identity) + reference video (motion) -> animated video.
+dataset_snapshot_download(
+    "DiffSynth-Studio/diffsynth_example_dataset",
+    local_dir="data/diffsynth_example_dataset",
+    allow_file_pattern="wanvideo/Wan2.2-Animate-2-14B/*"
+)
+reference_image = Image.open("data/diffsynth_example_dataset/wanvideo/Wan2.2-Animate-2-14B/refimage.jpg").convert("RGB")
+reference_video = VideoData("data/diffsynth_example_dataset/wanvideo/Wan2.2-Animate-2-14B/refvideo.mp4").raw_data()
+# Example 1: single-clip generation
+num_frames = 81
+video = pipe(
+    prompt="人物外观描述：一名长黑发女性，穿着白色半透明蕾丝长袖上衣，衣身带有花卉刺绣，下身搭配白色百褶短裙和黑色腰带，脚穿米白色厚底运动鞋。 背景描述：背景为现代室内空间，墙面和柜体以浅灰色为主，后方设有两扇深色落地窗或玻璃门，顶部安装长条形灯具，中央有一块浅色长方形台面。",
+    negative_prompt="色调艳丽，过曝，静态，细节模糊不清，字幕，风格，作品，画作，画面，静止，整体发灰，最差质量，低质量，JPEG压缩残留，丑陋的，残缺的，多余的手指，画得不好的手部，画得不好的脸部，畸形的，毁容的，形态畸形的肢体，手指融合，静止不动的画面，杂乱的背景，三条腿，背景人很多，倒着走",
+    animate2_prompt_ref="视频中的人在做动作，背景静止",
+    animate2_reference_image=reference_image,
+    animate2_reference_video=reference_video[:num_frames],
+    animate2_offload_kv=True,
+    num_frames=num_frames, height=1280, width=720,
+    num_inference_steps=40, cfg_scale=3.0,
+    seed=0, tiled=True,
+)
+if dist.get_rank() == 0:
+    save_video(video, "video_Wan2.2-Animate-2-14B.mp4", fps=24, quality=5)
+
+
+# Example 2: multi-clip long-video generation
+def generate_long_video(pipe, reference_image, cond_images, clip_len, first_num=1, **kwargs):
+    assert clip_len > first_num, "clip_len must be greater than first_num"
+
+    def zigzag_padding(array, target_len):
+        if len(array) == 1:
+            return [array[0]] * target_len
+        idx, flip, out = 0, False, []
+        while len(out) < target_len:
+            out.append(array[idx])
+            idx += -1 if flip else 1
+            if idx == 0 or idx == len(array) - 1:
+                flip = not flip
+        return out[:target_len]
+
+    real_len = len(cond_images)
+    if real_len == 0:
+        return []
+    step = clip_len - first_num
+    # Precompute clip count so clips of `clip_len` stepping by `step` tile the (padded) driving video.
+    num_clips = 1 if real_len <= clip_len else (real_len - clip_len + step - 1) // step + 1
+    target_len = clip_len + (num_clips - 1) * step
+    if real_len < target_len:
+        cond_images = zigzag_padding(cond_images, target_len)
+
+    all_frames = []
+    prev_tail = None
+    for i in range(num_clips):
+        start = i * step
+        seg_driving = cond_images[start:start + clip_len]
+        seg_out = pipe(
+            animate2_reference_image=reference_image,
+            animate2_reference_video=seg_driving,
+            animate2_refert_images=None if i == 0 else prev_tail,
+            num_frames=clip_len,
+            **kwargs,
+        )
+        prev_tail = seg_out[-first_num:]
+        if i != 0:
+            seg_out = seg_out[first_num:]
+        all_frames.extend(seg_out)
+    return all_frames[:real_len]
+
+
+clip_len = 81
+long_video = generate_long_video(
+    pipe,
+    reference_image=reference_image,
+    cond_images=reference_video,
+    clip_len=clip_len,
+    first_num=1,
+    prompt="人物外观描述：一名长黑发女性，穿着白色半透明蕾丝长袖上衣，衣身带有花卉刺绣，下身搭配白色百褶短裙和黑色腰带，脚穿米白色厚底运动鞋。 背景描述：背景为现代室内空间，墙面和柜体以浅灰色为主，后方设有两扇深色落地窗或玻璃门，顶部安装长条形灯具，中央有一块浅色长方形台面。",
+    negative_prompt="色调艳丽，过曝，静态，细节模糊不清，字幕，风格，作品，画作，画面，静止，整体发灰，最差质量，低质量，JPEG压缩残留，丑陋的，残缺的，多余的手指，画得不好的手部，画得不好的脸部，畸形的，毁容的，形态畸形的肢体，手指融合，静止不动的画面，杂乱的背景，三条腿，背景人很多，倒着走",
+    animate2_prompt_ref="视频中的人在做动作，背景静止",
+    animate2_offload_kv=True,
+    height=1280, width=720,
+    num_inference_steps=40, cfg_scale=3.0,
+    seed=0, tiled=True,
+)
+if dist.get_rank() == 0:
+    save_video(long_video, "video_Wan2.2-Animate-2-14B-long.mp4", fps=24, quality=5)
diff --git a/examples/wanvideo/model_inference/Wan2.2-Animate-2-14B-Distilled.py b/examples/wanvideo/model_inference/Wan2.2-Animate-2-14B-Distilled.py
new file mode 100644
index 000000000..2ed119918
--- /dev/null
+++ b/examples/wanvideo/model_inference/Wan2.2-Animate-2-14B-Distilled.py
@@ -0,0 +1,115 @@
+import torch
+from PIL import Image
+from diffsynth.utils.data import save_video, VideoData
+from diffsynth.pipelines.wan_video import WanVideoPipeline, ModelConfig
+from modelscope import dataset_snapshot_download
+
+vram_config = {
+    "offload_dtype": torch.bfloat16,
+    "offload_device": "cpu",
+    "onload_dtype": torch.bfloat16,
+    "onload_device": "cuda",
+    "preparing_dtype": torch.bfloat16,
+    "preparing_device": "cuda",
+    "computation_dtype": torch.bfloat16,
+    "computation_device": "cuda",
+}
+
+pipe = WanVideoPipeline.from_pretrained(
+    torch_dtype=torch.bfloat16,
+    device="cuda",
+    model_configs=[
+        ModelConfig(model_id="Wan-AI/Wan2.2-Animate-2-14B", origin_file_pattern="wan_animate_2/wan_animate_2_bf16_distillation.safetensors", **vram_config),
+        ModelConfig(model_id="Wan-AI/Wan2.2-Animate-2-14B", origin_file_pattern="videomodel/Wan-AI/models_t5_umt5-xxl-enc-bf16.pth", **vram_config),
+        ModelConfig(model_id="Wan-AI/Wan2.2-Animate-2-14B", origin_file_pattern="videomodel/Wan-AI/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth", **vram_config),
+        ModelConfig(model_id="Wan-AI/Wan2.1-T2V-14B", origin_file_pattern="Wan2.1_VAE.pth", **vram_config),
+    ],
+    tokenizer_config=ModelConfig(model_id="Wan-AI/Wan2.2-Animate-2-14B", origin_file_pattern="videomodel/Wan-AI/umt5-xxl/"),
+)
+
+# Character animation: reference image (identity) + reference video (motion) -> animated video.
+dataset_snapshot_download(
+    "DiffSynth-Studio/diffsynth_example_dataset",
+    local_dir="data/diffsynth_example_dataset",
+    allow_file_pattern="wanvideo/Wan2.2-Animate-2-14B-Distilled/*"
+)
+reference_image = Image.open("data/diffsynth_example_dataset/wanvideo/Wan2.2-Animate-2-14B-Distilled/refimage.jpg").convert("RGB")
+reference_video = VideoData("data/diffsynth_example_dataset/wanvideo/Wan2.2-Animate-2-14B-Distilled/refvideo.mp4").raw_data()
+
+# Example 1: single-clip generation
+num_frames = 81
+# For distilled model, set animate2_log_scale to -1.3, num_inference_steps to 10, and cfg_scale to 1.0.
+video = pipe(
+    prompt="人物外观描述：一名长黑发女性，穿着白色半透明蕾丝长袖上衣，衣身带有花卉刺绣，下身搭配白色百褶短裙和黑色腰带，脚穿米白色厚底运动鞋。 背景描述：背景为现代室内空间，墙面和柜体以浅灰色为主，后方设有两扇深色落地窗或玻璃门，顶部安装长条形灯具，中央有一块浅色长方形台面。",
+    animate2_prompt_ref="视频中的人在做动作，背景静止",
+    animate2_reference_image=reference_image,
+    animate2_reference_video=reference_video[:num_frames],
+    animate2_offload_kv=True,
+    animate2_log_scale=-1.3,
+    num_frames=num_frames, height=1280, width=720,
+    num_inference_steps=10, cfg_scale=1.0,
+    seed=0, tiled=True,
+)
+save_video(video, "video_Wan2.2-Animate-2-14B-Distilled.mp4", fps=24, quality=5)
+
+
+# Example 2: multi-clip long-video generation
+def generate_long_video(pipe, reference_image, cond_images, clip_len, first_num=1, **kwargs):
+    assert clip_len > first_num, "clip_len must be greater than first_num"
+
+    def zigzag_padding(array, target_len):
+        if len(array) == 1:
+            return [array[0]] * target_len
+        idx, flip, out = 0, False, []
+        while len(out) < target_len:
+            out.append(array[idx])
+            idx += -1 if flip else 1
+            if idx == 0 or idx == len(array) - 1:
+                flip = not flip
+        return out[:target_len]
+
+    real_len = len(cond_images)
+    if real_len == 0:
+        return []
+    step = clip_len - first_num
+    # Precompute clip count so clips of `clip_len` stepping by `step` tile the (padded) driving video.
+    num_clips = 1 if real_len <= clip_len else (real_len - clip_len + step - 1) // step + 1
+    target_len = clip_len + (num_clips - 1) * step
+    if real_len < target_len:
+        cond_images = zigzag_padding(cond_images, target_len)
+
+    all_frames = []
+    prev_tail = None
+    for i in range(num_clips):
+        start = i * step
+        seg_driving = cond_images[start:start + clip_len]
+        seg_out = pipe(
+            animate2_reference_image=reference_image,
+            animate2_reference_video=seg_driving,
+            animate2_refert_images=None if i == 0 else prev_tail,
+            num_frames=clip_len,
+            **kwargs,
+        )
+        prev_tail = seg_out[-first_num:]
+        if i != 0:
+            seg_out = seg_out[first_num:]
+        all_frames.extend(seg_out)
+    return all_frames[:real_len]
+
+
+# For distilled model, set animate2_log_scale to -1.3, num_inference_steps to 10, and cfg_scale to 1.0.
+long_video = generate_long_video(
+    pipe,
+    reference_image=reference_image,
+    cond_images=reference_video,
+    clip_len=81,
+    first_num=1,
+    prompt="人物外观描述：一名长黑发女性，穿着白色半透明蕾丝长袖上衣，衣身带有花卉刺绣，下身搭配白色百褶短裙和黑色腰带，脚穿米白色厚底运动鞋。 背景描述：背景为现代室内空间，墙面和柜体以浅灰色为主，后方设有两扇深色落地窗或玻璃门，顶部安装长条形灯具，中央有一块浅色长方形台面。",
+    animate2_prompt_ref="视频中的人在做动作，背景静止",
+    animate2_offload_kv=True,
+    animate2_log_scale=-1.3,
+    height=1280, width=720,
+    num_inference_steps=10, cfg_scale=1.0,
+    seed=0, tiled=True,
+)
+save_video(long_video, "video_Wan2.2-Animate-2-14B-Distilled-long.mp4", fps=24, quality=5)
diff --git a/examples/wanvideo/model_inference/Wan2.2-Animate-2-14B.py b/examples/wanvideo/model_inference/Wan2.2-Animate-2-14B.py
new file mode 100644
index 000000000..b74e56da3
--- /dev/null
+++ b/examples/wanvideo/model_inference/Wan2.2-Animate-2-14B.py
@@ -0,0 +1,113 @@
+import torch
+from PIL import Image
+from diffsynth.utils.data import save_video, VideoData
+from diffsynth.pipelines.wan_video import WanVideoPipeline, ModelConfig
+from modelscope import dataset_snapshot_download
+
+vram_config = {
+    "offload_dtype": torch.bfloat16,
+    "offload_device": "cpu",
+    "onload_dtype": torch.bfloat16,
+    "onload_device": "cuda",
+    "preparing_dtype": torch.bfloat16,
+    "preparing_device": "cuda",
+    "computation_dtype": torch.bfloat16,
+    "computation_device": "cuda",
+}
+
+pipe = WanVideoPipeline.from_pretrained(
+    torch_dtype=torch.bfloat16,
+    device="cuda",
+    model_configs=[
+        ModelConfig(model_id="Wan-AI/Wan2.2-Animate-2-14B", origin_file_pattern="wan_animate_2/wan_animate_2_bf16.safetensors", **vram_config),
+        ModelConfig(model_id="Wan-AI/Wan2.2-Animate-2-14B", origin_file_pattern="videomodel/Wan-AI/models_t5_umt5-xxl-enc-bf16.pth", **vram_config),
+        ModelConfig(model_id="Wan-AI/Wan2.2-Animate-2-14B", origin_file_pattern="videomodel/Wan-AI/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth", **vram_config),
+        ModelConfig(model_id="Wan-AI/Wan2.1-T2V-14B", origin_file_pattern="Wan2.1_VAE.pth", **vram_config),
+    ],
+    tokenizer_config=ModelConfig(model_id="Wan-AI/Wan2.2-Animate-2-14B", origin_file_pattern="videomodel/Wan-AI/umt5-xxl/"),
+)
+
+# Character animation: reference image (identity) + reference video (motion) -> animated video.
+dataset_snapshot_download(
+    "DiffSynth-Studio/diffsynth_example_dataset",
+    local_dir="data/diffsynth_example_dataset",
+    allow_file_pattern="wanvideo/Wan2.2-Animate-2-14B/*"
+)
+reference_image = Image.open("data/diffsynth_example_dataset/wanvideo/Wan2.2-Animate-2-14B/refimage.jpg").convert("RGB")
+reference_video = VideoData("data/diffsynth_example_dataset/wanvideo/Wan2.2-Animate-2-14B/refvideo.mp4").raw_data()
+# Example 1: single-clip generation
+num_frames = 81
+video = pipe(
+    prompt="人物外观描述：一名长黑发女性，穿着白色半透明蕾丝长袖上衣，衣身带有花卉刺绣，下身搭配白色百褶短裙和黑色腰带，脚穿米白色厚底运动鞋。 背景描述：背景为现代室内空间，墙面和柜体以浅灰色为主，后方设有两扇深色落地窗或玻璃门，顶部安装长条形灯具，中央有一块浅色长方形台面。",
+    negative_prompt="色调艳丽，过曝，静态，细节模糊不清，字幕，风格，作品，画作，画面，静止，整体发灰，最差质量，低质量，JPEG压缩残留，丑陋的，残缺的，多余的手指，画得不好的手部，画得不好的脸部，畸形的，毁容的，形态畸形的肢体，手指融合，静止不动的画面，杂乱的背景，三条腿，背景人很多，倒着走",
+    animate2_prompt_ref="视频中的人在做动作，背景静止",
+    animate2_reference_image=reference_image,
+    animate2_reference_video=reference_video[:num_frames],
+    animate2_offload_kv=True,
+    num_frames=num_frames, height=1280, width=720,
+    num_inference_steps=40, cfg_scale=3.0,
+    seed=0, tiled=True,
+)
+save_video(video, "video_Wan2.2-Animate-2-14B.mp4", fps=24, quality=5)
+
+
+# Example 2: multi-clip long-video generation
+def generate_long_video(pipe, reference_image, cond_images, clip_len, first_num=1, **kwargs):
+    assert clip_len > first_num, "clip_len must be greater than first_num"
+
+    def zigzag_padding(array, target_len):
+        if len(array) == 1:
+            return [array[0]] * target_len
+        idx, flip, out = 0, False, []
+        while len(out) < target_len:
+            out.append(array[idx])
+            idx += -1 if flip else 1
+            if idx == 0 or idx == len(array) - 1:
+                flip = not flip
+        return out[:target_len]
+
+    real_len = len(cond_images)
+    if real_len == 0:
+        return []
+    step = clip_len - first_num
+    # Precompute clip count so clips of `clip_len` stepping by `step` tile the (padded) driving video.
+    num_clips = 1 if real_len <= clip_len else (real_len - clip_len + step - 1) // step + 1
+    target_len = clip_len + (num_clips - 1) * step
+    if real_len < target_len:
+        cond_images = zigzag_padding(cond_images, target_len)
+
+    all_frames = []
+    prev_tail = None
+    for i in range(num_clips):
+        start = i * step
+        seg_driving = cond_images[start:start + clip_len]
+        seg_out = pipe(
+            animate2_reference_image=reference_image,
+            animate2_reference_video=seg_driving,
+            animate2_refert_images=None if i == 0 else prev_tail,
+            num_frames=clip_len,
+            **kwargs,
+        )
+        prev_tail = seg_out[-first_num:]
+        if i != 0:
+            seg_out = seg_out[first_num:]
+        all_frames.extend(seg_out)
+    return all_frames[:real_len]
+
+
+clip_len = 81
+long_video = generate_long_video(
+    pipe,
+    reference_image=reference_image,
+    cond_images=reference_video,
+    clip_len=clip_len,
+    first_num=1,
+    prompt="人物外观描述：一名长黑发女性，穿着白色半透明蕾丝长袖上衣，衣身带有花卉刺绣，下身搭配白色百褶短裙和黑色腰带，脚穿米白色厚底运动鞋。 背景描述：背景为现代室内空间，墙面和柜体以浅灰色为主，后方设有两扇深色落地窗或玻璃门，顶部安装长条形灯具，中央有一块浅色长方形台面。",
+    negative_prompt="色调艳丽，过曝，静态，细节模糊不清，字幕，风格，作品，画作，画面，静止，整体发灰，最差质量，低质量，JPEG压缩残留，丑陋的，残缺的，多余的手指，画得不好的手部，画得不好的脸部，畸形的，毁容的，形态畸形的肢体，手指融合，静止不动的画面，杂乱的背景，三条腿，背景人很多，倒着走",
+    animate2_prompt_ref="视频中的人在做动作，背景静止",
+    animate2_offload_kv=True,
+    height=1280, width=720,
+    num_inference_steps=40, cfg_scale=3.0,
+    seed=0, tiled=True,
+)
+save_video(long_video, "video_Wan2.2-Animate-2-14B-long.mp4", fps=24, quality=5)
diff --git a/examples/wanvideo/model_inference_low_vram/Wan2.2-Animate-2-14B-Distilled.py b/examples/wanvideo/model_inference_low_vram/Wan2.2-Animate-2-14B-Distilled.py
new file mode 100644
index 000000000..0d4e5aa5d
--- /dev/null
+++ b/examples/wanvideo/model_inference_low_vram/Wan2.2-Animate-2-14B-Distilled.py
@@ -0,0 +1,117 @@
+import torch
+from PIL import Image
+from diffsynth.utils.data import save_video, VideoData
+from diffsynth.pipelines.wan_video import WanVideoPipeline, ModelConfig
+from modelscope import dataset_snapshot_download
+
+vram_config = {
+    "offload_dtype": "disk",
+    "offload_device": "disk",
+    "onload_dtype": torch.bfloat16,
+    "onload_device": "cpu",
+    "preparing_dtype": torch.bfloat16,
+    "preparing_device": "cuda",
+    "computation_dtype": torch.bfloat16,
+    "computation_device": "cuda",
+}
+
+pipe = WanVideoPipeline.from_pretrained(
+    torch_dtype=torch.bfloat16,
+    device="cuda",
+    model_configs=[
+        ModelConfig(model_id="Wan-AI/Wan2.2-Animate-2-14B", origin_file_pattern="wan_animate_2/wan_animate_2_bf16_distillation.safetensors", **vram_config),
+        ModelConfig(model_id="Wan-AI/Wan2.2-Animate-2-14B", origin_file_pattern="videomodel/Wan-AI/models_t5_umt5-xxl-enc-bf16.pth", **vram_config),
+        ModelConfig(model_id="Wan-AI/Wan2.2-Animate-2-14B", origin_file_pattern="videomodel/Wan-AI/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth", **vram_config),
+        ModelConfig(model_id="Wan-AI/Wan2.1-T2V-14B", origin_file_pattern="Wan2.1_VAE.pth", **vram_config),
+    ],
+    tokenizer_config=ModelConfig(model_id="Wan-AI/Wan2.2-Animate-2-14B", origin_file_pattern="videomodel/Wan-AI/umt5-xxl/"),
+    vram_limit=torch.cuda.mem_get_info("cuda")[1] / (1024 ** 3) - 0.5,
+)
+
+# Character animation: reference image (identity) + reference video (motion) -> animated video.
+dataset_snapshot_download(
+    "DiffSynth-Studio/diffsynth_example_dataset",
+    local_dir="data/diffsynth_example_dataset",
+    allow_file_pattern="wanvideo/Wan2.2-Animate-2-14B-Distilled/*"
+)
+reference_image = Image.open("data/diffsynth_example_dataset/wanvideo/Wan2.2-Animate-2-14B-Distilled/refimage.jpg").convert("RGB")
+reference_video = VideoData("data/diffsynth_example_dataset/wanvideo/Wan2.2-Animate-2-14B-Distilled/refvideo.mp4").raw_data()
+
+# Example 1: single-clip generation
+num_frames = 81
+# For distilled model, set animate2_log_scale to -1.3, num_inference_steps to 10, and cfg_scale to 1.0.
+video = pipe(
+    prompt="人物外观描述：一名长黑发女性，穿着白色半透明蕾丝长袖上衣，衣身带有花卉刺绣，下身搭配白色百褶短裙和黑色腰带，脚穿米白色厚底运动鞋。 背景描述：背景为现代室内空间，墙面和柜体以浅灰色为主，后方设有两扇深色落地窗或玻璃门，顶部安装长条形灯具，中央有一块浅色长方形台面。",
+    animate2_prompt_ref="视频中的人在做动作，背景静止",
+    animate2_reference_image=reference_image,
+    animate2_reference_video=reference_video[:num_frames],
+    animate2_offload_kv=True,
+    animate2_log_scale=-1.3,
+    num_frames=num_frames, height=1280, width=720,
+    num_inference_steps=10, cfg_scale=1.0,
+    seed=0, tiled=True,
+)
+save_video(video, "video_Wan2.2-Animate-2-14B-Distilled.mp4", fps=24, quality=5)
+
+
+# Example 2: multi-clip long-video generation
+def generate_long_video(pipe, reference_image, cond_images, clip_len, first_num=1, **kwargs):
+    assert clip_len > first_num, "clip_len must be greater than first_num"
+
+    def zigzag_padding(array, target_len):
+        if len(array) == 1:
+            return [array[0]] * target_len
+        idx, flip, out = 0, False, []
+        while len(out) < target_len:
+            out.append(array[idx])
+            idx += -1 if flip else 1
+            if idx == 0 or idx == len(array) - 1:
+                flip = not flip
+        return out[:target_len]
+
+    real_len = len(cond_images)
+    if real_len == 0:
+        return []
+    step = clip_len - first_num
+    # Precompute clip count so clips of `clip_len` stepping by `step` tile the (padded) driving video.
+    num_clips = 1 if real_len <= clip_len else (real_len - clip_len + step - 1) // step + 1
+    target_len = clip_len + (num_clips - 1) * step
+    if real_len < target_len:
+        cond_images = zigzag_padding(cond_images, target_len)
+
+    all_frames = []
+    prev_tail = None
+    for i in range(num_clips):
+        start = i * step
+        seg_driving = cond_images[start:start + clip_len]
+        seg_out = pipe(
+            animate2_reference_image=reference_image,
+            animate2_reference_video=seg_driving,
+            animate2_refert_images=None if i == 0 else prev_tail,
+            num_frames=clip_len,
+            **kwargs,
+        )
+        prev_tail = seg_out[-first_num:]
+        if i != 0:
+            seg_out = seg_out[first_num:]
+        all_frames.extend(seg_out)
+    return all_frames[:real_len]
+
+
+clip_len = 81
+# For distilled model, set animate2_log_scale to -1.3, num_inference_steps to 10, and cfg_scale to 1.0.
+long_video = generate_long_video(
+    pipe,
+    reference_image=reference_image,
+    cond_images=reference_video,
+    clip_len=clip_len,
+    first_num=1,
+    prompt="人物外观描述：一名长黑发女性，穿着白色半透明蕾丝长袖上衣，衣身带有花卉刺绣，下身搭配白色百褶短裙和黑色腰带，脚穿米白色厚底运动鞋。 背景描述：背景为现代室内空间，墙面和柜体以浅灰色为主，后方设有两扇深色落地窗或玻璃门，顶部安装长条形灯具，中央有一块浅色长方形台面。",
+    animate2_prompt_ref="视频中的人在做动作，背景静止",
+    animate2_offload_kv=True,
+    animate2_log_scale=-1.3,
+    height=1280, width=720,
+    num_inference_steps=10, cfg_scale=1.0,
+    seed=0, tiled=True,
+)
+save_video(long_video, "video_Wan2.2-Animate-2-14B-Distilled-long.mp4", fps=24, quality=5)
diff --git a/examples/wanvideo/model_inference_low_vram/Wan2.2-Animate-2-14B.py b/examples/wanvideo/model_inference_low_vram/Wan2.2-Animate-2-14B.py
new file mode 100644
index 000000000..d89a51294
--- /dev/null
+++ b/examples/wanvideo/model_inference_low_vram/Wan2.2-Animate-2-14B.py
@@ -0,0 +1,114 @@
+import torch
+from PIL import Image
+from diffsynth.utils.data import save_video, VideoData
+from diffsynth.pipelines.wan_video import WanVideoPipeline, ModelConfig
+from modelscope import dataset_snapshot_download
+
+vram_config = {
+    "offload_dtype": "disk",
+    "offload_device": "disk",
+    "onload_dtype": torch.bfloat16,
+    "onload_device": "cpu",
+    "preparing_dtype": torch.bfloat16,
+    "preparing_device": "cuda",
+    "computation_dtype": torch.bfloat16,
+    "computation_device": "cuda",
+}
+
+pipe = WanVideoPipeline.from_pretrained(
+    torch_dtype=torch.bfloat16,
+    device="cuda",
+    model_configs=[
+        ModelConfig(model_id="Wan-AI/Wan2.2-Animate-2-14B", origin_file_pattern="wan_animate_2/wan_animate_2_bf16.safetensors", **vram_config),
+        ModelConfig(model_id="Wan-AI/Wan2.2-Animate-2-14B", origin_file_pattern="videomodel/Wan-AI/models_t5_umt5-xxl-enc-bf16.pth", **vram_config),
+        ModelConfig(model_id="Wan-AI/Wan2.2-Animate-2-14B", origin_file_pattern="videomodel/Wan-AI/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth", **vram_config),
+        ModelConfig(model_id="Wan-AI/Wan2.1-T2V-14B", origin_file_pattern="Wan2.1_VAE.pth", **vram_config),
+    ],
+    tokenizer_config=ModelConfig(model_id="Wan-AI/Wan2.2-Animate-2-14B", origin_file_pattern="videomodel/Wan-AI/umt5-xxl/"),
+    vram_limit=torch.cuda.mem_get_info("cuda")[1] / (1024 ** 3) - 0.5,
+)
+
+# Character animation: reference image (identity) + reference video (motion) -> animated video.
+dataset_snapshot_download(
+    "DiffSynth-Studio/diffsynth_example_dataset",
+    local_dir="data/diffsynth_example_dataset",
+    allow_file_pattern="wanvideo/Wan2.2-Animate-2-14B/*"
+)
+reference_image = Image.open("data/diffsynth_example_dataset/wanvideo/Wan2.2-Animate-2-14B/refimage.jpg").convert("RGB")
+reference_video = VideoData("data/diffsynth_example_dataset/wanvideo/Wan2.2-Animate-2-14B/refvideo.mp4").raw_data()
+# Example 1: single-clip generation
+num_frames = 81
+video = pipe(
+    prompt="人物外观描述：一名长黑发女性，穿着白色半透明蕾丝长袖上衣，衣身带有花卉刺绣，下身搭配白色百褶短裙和黑色腰带，脚穿米白色厚底运动鞋。 背景描述：背景为现代室内空间，墙面和柜体以浅灰色为主，后方设有两扇深色落地窗或玻璃门，顶部安装长条形灯具，中央有一块浅色长方形台面。",
+    negative_prompt="色调艳丽，过曝，静态，细节模糊不清，字幕，风格，作品，画作，画面，静止，整体发灰，最差质量，低质量，JPEG压缩残留，丑陋的，残缺的，多余的手指，画得不好的手部，画得不好的脸部，畸形的，毁容的，形态畸形的肢体，手指融合，静止不动的画面，杂乱的背景，三条腿，背景人很多，倒着走",
+    animate2_prompt_ref="视频中的人在做动作，背景静止",
+    animate2_reference_image=reference_image,
+    animate2_reference_video=reference_video[:num_frames],
+    animate2_offload_kv=True,
+    num_frames=num_frames, height=1280, width=720,
+    num_inference_steps=40, cfg_scale=3.0,
+    seed=0, tiled=True,
+)
+save_video(video, "video_Wan2.2-Animate-2-14B.mp4", fps=24, quality=5)
+
+
+# Example 2: multi-clip long-video generation
+def generate_long_video(pipe, reference_image, cond_images, clip_len, first_num=1, **kwargs):
+    assert clip_len > first_num, "clip_len must be greater than first_num"
+
+    def zigzag_padding(array, target_len):
+        if len(array) == 1:
+            return [array[0]] * target_len
+        idx, flip, out = 0, False, []
+        while len(out) < target_len:
+            out.append(array[idx])
+            idx += -1 if flip else 1
+            if idx == 0 or idx == len(array) - 1:
+                flip = not flip
+        return out[:target_len]
+
+    real_len = len(cond_images)
+    if real_len == 0:
+        return []
+    step = clip_len - first_num
+    # Precompute clip count so clips of `clip_len` stepping by `step` tile the (padded) driving video.
+    num_clips = 1 if real_len <= clip_len else (real_len - clip_len + step - 1) // step + 1
+    target_len = clip_len + (num_clips - 1) * step
+    if real_len < target_len:
+        cond_images = zigzag_padding(cond_images, target_len)
+
+    all_frames = []
+    prev_tail = None
+    for i in range(num_clips):
+        start = i * step
+        seg_driving = cond_images[start:start + clip_len]
+        seg_out = pipe(
+            animate2_reference_image=reference_image,
+            animate2_reference_video=seg_driving,
+            animate2_refert_images=None if i == 0 else prev_tail,
+            num_frames=clip_len,
+            **kwargs,
+        )
+        prev_tail = seg_out[-first_num:]
+        if i != 0:
+            seg_out = seg_out[first_num:]
+        all_frames.extend(seg_out)
+    return all_frames[:real_len]
+
+
+clip_len = 81
+long_video = generate_long_video(
+    pipe,
+    reference_image=reference_image,
+    cond_images=reference_video,
+    clip_len=clip_len,
+    first_num=1,
+    prompt="人物外观描述：一名长黑发女性，穿着白色半透明蕾丝长袖上衣，衣身带有花卉刺绣，下身搭配白色百褶短裙和黑色腰带，脚穿米白色厚底运动鞋。 背景描述：背景为现代室内空间，墙面和柜体以浅灰色为主，后方设有两扇深色落地窗或玻璃门，顶部安装长条形灯具，中央有一块浅色长方形台面。",
+    negative_prompt="色调艳丽，过曝，静态，细节模糊不清，字幕，风格，作品，画作，画面，静止，整体发灰，最差质量，低质量，JPEG压缩残留，丑陋的，残缺的，多余的手指，画得不好的手部，画得不好的脸部，畸形的，毁容的，形态畸形的肢体，手指融合，静止不动的画面，杂乱的背景，三条腿，背景人很多，倒着走",
+    animate2_prompt_ref="视频中的人在做动作，背景静止",
+    animate2_offload_kv=True,
+    height=1280, width=720,
+    num_inference_steps=40, cfg_scale=3.0,
+    seed=0, tiled=True,
+)
+save_video(long_video, "video_Wan2.2-Animate-2-14B-long.mp4", fps=24, quality=5)
diff --git a/examples/wanvideo/model_training/full/Wan2.2-Animate-2-14B-Distilled.sh b/examples/wanvideo/model_training/full/Wan2.2-Animate-2-14B-Distilled.sh
new file mode 100644
index 000000000..fdc450268
--- /dev/null
+++ b/examples/wanvideo/model_training/full/Wan2.2-Animate-2-14B-Distilled.sh
@@ -0,0 +1,35 @@
+modelscope download --dataset DiffSynth-Studio/diffsynth_example_dataset --include "wanvideo/Wan2.2-Animate-2-14B-Distilled/*" --local_dir ./data/diffsynth_example_dataset
+
+accelerate launch examples/wanvideo/model_training/train.py \
+  --dataset_base_path data/diffsynth_example_dataset/wanvideo/Wan2.2-Animate-2-14B-Distilled \
+  --dataset_metadata_path data/diffsynth_example_dataset/wanvideo/Wan2.2-Animate-2-14B-Distilled/metadata.json \
+  --data_file_keys "video,animate2_reference_image,animate2_reference_video" \
+  --height 640 \
+  --width 352 \
+  --num_frames 81 \
+  --dataset_repeat 1 \
+  --model_id_with_origin_paths "Wan-AI/Wan2.2-Animate-2-14B:videomodel/Wan-AI/models_t5_umt5-xxl-enc-bf16.pth,Wan-AI/Wan2.2-Animate-2-14B:videomodel/Wan-AI/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth,Wan-AI/Wan2.1-T2V-14B:Wan2.1_VAE.pth" \
+  --learning_rate 1e-5 \
+  --num_epochs 2 \
+  --remove_prefix_in_ckpt "pipe.dit." \
+  --output_path "./models/train/Wan2.2-Animate-2-14B-Distilled_full_splited_cache" \
+  --extra_inputs "animate2_prompt_ref,animate2_reference_image,animate2_reference_video" \
+  --use_gradient_checkpointing \
+  --task "sft:data_process"
+
+accelerate launch --config_file examples/wanvideo/model_training/full/accelerate_config_14B.yaml examples/wanvideo/model_training/train.py \
+  --dataset_base_path models/train/Wan2.2-Animate-2-14B-Distilled_full_splited_cache \
+  --data_file_keys "video,animate2_reference_image,animate2_reference_video" \
+  --height 640 \
+  --width 352 \
+  --num_frames 81 \
+  --dataset_repeat 100 \
+  --model_id_with_origin_paths "Wan-AI/Wan2.2-Animate-2-14B:wan_animate_2/wan_animate_2_bf16_distillation.safetensors" \
+  --learning_rate 1e-5 \
+  --num_epochs 2 \
+  --remove_prefix_in_ckpt "pipe.dit." \
+  --output_path "./models/train/Wan2.2-Animate-2-14B-Distilled_full" \
+  --trainable_models "dit.blocks.0.block.self_attn.q,dit.blocks.0.block.self_attn.k,dit.blocks.0.block.self_attn.v,dit.blocks.1.block.self_attn.q,dit.blocks.1.block.self_attn.k,dit.blocks.1.block.self_attn.v,dit.blocks.2.block.self_attn.q,dit.blocks.2.block.self_attn.k,dit.blocks.2.block.self_attn.v,dit.blocks.3.block.self_attn.q,dit.blocks.3.block.self_attn.k,dit.blocks.3.block.self_attn.v,dit.blocks.4.block.self_attn.q,dit.blocks.4.block.self_attn.k,dit.blocks.4.block.self_attn.v,dit.blocks.5.block.self_attn.q,dit.blocks.5.block.self_attn.k,dit.blocks.5.block.self_attn.v,dit.blocks.6.block.self_attn.q,dit.blocks.6.block.self_attn.k,dit.blocks.6.block.self_attn.v,dit.blocks.7.block.self_attn.q,dit.blocks.7.block.self_attn.k,dit.blocks.7.block.self_attn.v,dit.blocks.8.block.self_attn.q,dit.blocks.8.block.self_attn.k,dit.blocks.8.block.self_attn.v,dit.blocks.9.block.self_attn.q,dit.blocks.9.block.self_attn.k,dit.blocks.9.block.self_attn.v,dit.blocks.10.block.self_attn.q,dit.blocks.10.block.self_attn.k,dit.blocks.10.block.self_attn.v,dit.blocks.11.block.self_attn.q,dit.blocks.11.block.self_attn.k,dit.blocks.11.block.self_attn.v,dit.blocks.12.block.self_attn.q,dit.blocks.12.block.self_attn.k,dit.blocks.12.block.self_attn.v,dit.blocks.13.block.self_attn.q,dit.blocks.13.block.self_attn.k,dit.blocks.13.block.self_attn.v,dit.blocks.14.block.self_attn.q,dit.blocks.14.block.self_attn.k,dit.blocks.14.block.self_attn.v,dit.blocks.15.block.self_attn.q,dit.blocks.15.block.self_attn.k,dit.blocks.15.block.self_attn.v,dit.blocks.16.block.self_attn.q,dit.blocks.16.block.self_attn.k,dit.blocks.16.block.self_attn.v,dit.blocks.17.block.self_attn.q,dit.blocks.17.block.self_attn.k,dit.blocks.17.block.self_attn.v,dit.blocks.18.block.self_attn.q,dit.blocks.18.block.self_attn.k,dit.blocks.18.block.self_attn.v,dit.blocks.19.block.self_attn.q,dit.blocks.19.block.self_attn.k,dit.blocks.19.block.self_attn.v,dit.blocks.20.block.self_attn.q,dit.blocks.20.block.self_attn.k,dit.blocks.20.block.self_attn.v,dit.blocks.21.block.self_attn.q,dit.blocks.21.block.self_attn.k,dit.blocks.21.block.self_attn.v,dit.blocks.22.block.self_attn.q,dit.blocks.22.block.self_attn.k,dit.blocks.22.block.self_attn.v,dit.blocks.23.block.self_attn.q,dit.blocks.23.block.self_attn.k,dit.blocks.23.block.self_attn.v,dit.blocks.24.block.self_attn.q,dit.blocks.24.block.self_attn.k,dit.blocks.24.block.self_attn.v,dit.blocks.25.block.self_attn.q,dit.blocks.25.block.self_attn.k,dit.blocks.25.block.self_attn.v,dit.blocks.26.block.self_attn.q,dit.blocks.26.block.self_attn.k,dit.blocks.26.block.self_attn.v,dit.blocks.27.block.self_attn.q,dit.blocks.27.block.self_attn.k,dit.blocks.27.block.self_attn.v,dit.blocks.28.block.self_attn.q,dit.blocks.28.block.self_attn.k,dit.blocks.28.block.self_attn.v,dit.blocks.29.block.self_attn.q,dit.blocks.29.block.self_attn.k,dit.blocks.29.block.self_attn.v,dit.blocks.30.block.self_attn.q,dit.blocks.30.block.self_attn.k,dit.blocks.30.block.self_attn.v,dit.blocks.31.block.self_attn.q,dit.blocks.31.block.self_attn.k,dit.blocks.31.block.self_attn.v,dit.blocks.32.block.self_attn.q,dit.blocks.32.block.self_attn.k,dit.blocks.32.block.self_attn.v,dit.blocks.33.block.self_attn.q,dit.blocks.33.block.self_attn.k,dit.blocks.33.block.self_attn.v,dit.blocks.34.block.self_attn.q,dit.blocks.34.block.self_attn.k,dit.blocks.34.block.self_attn.v,dit.blocks.35.block.self_attn.q,dit.blocks.35.block.self_attn.k,dit.blocks.35.block.self_attn.v,dit.blocks.36.block.self_attn.q,dit.blocks.36.block.self_attn.k,dit.blocks.36.block.self_attn.v,dit.blocks.37.block.self_attn.q,dit.blocks.37.block.self_attn.k,dit.blocks.37.block.self_attn.v,dit.blocks.38.block.self_attn.q,dit.blocks.38.block.self_attn.k,dit.blocks.38.block.self_attn.v,dit.blocks.39.block.self_attn.q,dit.blocks.39.block.self_attn.k,dit.blocks.39.block.self_attn.v" \
+  --extra_inputs "animate2_prompt_ref,animate2_reference_image,animate2_reference_video" \
+  --use_gradient_checkpointing \
+  --task "sft:train"
diff --git a/examples/wanvideo/model_training/full/Wan2.2-Animate-2-14B.sh b/examples/wanvideo/model_training/full/Wan2.2-Animate-2-14B.sh
new file mode 100644
index 000000000..c4fdf46b1
--- /dev/null
+++ b/examples/wanvideo/model_training/full/Wan2.2-Animate-2-14B.sh
@@ -0,0 +1,35 @@
+modelscope download --dataset DiffSynth-Studio/diffsynth_example_dataset --include "wanvideo/Wan2.2-Animate-2-14B/*" --local_dir ./data/diffsynth_example_dataset
+
+accelerate launch examples/wanvideo/model_training/train.py \
+  --dataset_base_path data/diffsynth_example_dataset/wanvideo/Wan2.2-Animate-2-14B \
+  --dataset_metadata_path data/diffsynth_example_dataset/wanvideo/Wan2.2-Animate-2-14B/metadata.json \
+  --data_file_keys "video,animate2_reference_image,animate2_reference_video" \
+  --height 640 \
+  --width 352 \
+  --num_frames 81 \
+  --dataset_repeat 1 \
+  --model_id_with_origin_paths "Wan-AI/Wan2.2-Animate-2-14B:videomodel/Wan-AI/models_t5_umt5-xxl-enc-bf16.pth,Wan-AI/Wan2.2-Animate-2-14B:videomodel/Wan-AI/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth,Wan-AI/Wan2.1-T2V-14B:Wan2.1_VAE.pth" \
+  --learning_rate 1e-5 \
+  --num_epochs 2 \
+  --remove_prefix_in_ckpt "pipe.dit." \
+  --output_path "./models/train/Wan2.2-Animate-2-14B_full_splited_cache" \
+  --extra_inputs "animate2_prompt_ref,animate2_reference_image,animate2_reference_video" \
+  --use_gradient_checkpointing \
+  --task "sft:data_process"
+
+accelerate launch --config_file examples/wanvideo/model_training/full/accelerate_config_14B.yaml examples/wanvideo/model_training/train.py \
+  --dataset_base_path models/train/Wan2.2-Animate-2-14B_full_splited_cache \
+  --data_file_keys "video,animate2_reference_image,animate2_reference_video" \
+  --height 640 \
+  --width 352 \
+  --num_frames 81 \
+  --dataset_repeat 100 \
+  --model_id_with_origin_paths "Wan-AI/Wan2.2-Animate-2-14B:wan_animate_2/wan_animate_2_bf16.safetensors" \
+  --learning_rate 1e-5 \
+  --num_epochs 2 \
+  --remove_prefix_in_ckpt "pipe.dit." \
+  --output_path "./models/train/Wan2.2-Animate-2-14B_full" \
+  --trainable_models "dit.blocks.0.block.self_attn.q,dit.blocks.0.block.self_attn.k,dit.blocks.0.block.self_attn.v,dit.blocks.1.block.self_attn.q,dit.blocks.1.block.self_attn.k,dit.blocks.1.block.self_attn.v,dit.blocks.2.block.self_attn.q,dit.blocks.2.block.self_attn.k,dit.blocks.2.block.self_attn.v,dit.blocks.3.block.self_attn.q,dit.blocks.3.block.self_attn.k,dit.blocks.3.block.self_attn.v,dit.blocks.4.block.self_attn.q,dit.blocks.4.block.self_attn.k,dit.blocks.4.block.self_attn.v,dit.blocks.5.block.self_attn.q,dit.blocks.5.block.self_attn.k,dit.blocks.5.block.self_attn.v,dit.blocks.6.block.self_attn.q,dit.blocks.6.block.self_attn.k,dit.blocks.6.block.self_attn.v,dit.blocks.7.block.self_attn.q,dit.blocks.7.block.self_attn.k,dit.blocks.7.block.self_attn.v,dit.blocks.8.block.self_attn.q,dit.blocks.8.block.self_attn.k,dit.blocks.8.block.self_attn.v,dit.blocks.9.block.self_attn.q,dit.blocks.9.block.self_attn.k,dit.blocks.9.block.self_attn.v,dit.blocks.10.block.self_attn.q,dit.blocks.10.block.self_attn.k,dit.blocks.10.block.self_attn.v,dit.blocks.11.block.self_attn.q,dit.blocks.11.block.self_attn.k,dit.blocks.11.block.self_attn.v,dit.blocks.12.block.self_attn.q,dit.blocks.12.block.self_attn.k,dit.blocks.12.block.self_attn.v,dit.blocks.13.block.self_attn.q,dit.blocks.13.block.self_attn.k,dit.blocks.13.block.self_attn.v,dit.blocks.14.block.self_attn.q,dit.blocks.14.block.self_attn.k,dit.blocks.14.block.self_attn.v,dit.blocks.15.block.self_attn.q,dit.blocks.15.block.self_attn.k,dit.blocks.15.block.self_attn.v,dit.blocks.16.block.self_attn.q,dit.blocks.16.block.self_attn.k,dit.blocks.16.block.self_attn.v,dit.blocks.17.block.self_attn.q,dit.blocks.17.block.self_attn.k,dit.blocks.17.block.self_attn.v,dit.blocks.18.block.self_attn.q,dit.blocks.18.block.self_attn.k,dit.blocks.18.block.self_attn.v,dit.blocks.19.block.self_attn.q,dit.blocks.19.block.self_attn.k,dit.blocks.19.block.self_attn.v,dit.blocks.20.block.self_attn.q,dit.blocks.20.block.self_attn.k,dit.blocks.20.block.self_attn.v,dit.blocks.21.block.self_attn.q,dit.blocks.21.block.self_attn.k,dit.blocks.21.block.self_attn.v,dit.blocks.22.block.self_attn.q,dit.blocks.22.block.self_attn.k,dit.blocks.22.block.self_attn.v,dit.blocks.23.block.self_attn.q,dit.blocks.23.block.self_attn.k,dit.blocks.23.block.self_attn.v,dit.blocks.24.block.self_attn.q,dit.blocks.24.block.self_attn.k,dit.blocks.24.block.self_attn.v,dit.blocks.25.block.self_attn.q,dit.blocks.25.block.self_attn.k,dit.blocks.25.block.self_attn.v,dit.blocks.26.block.self_attn.q,dit.blocks.26.block.self_attn.k,dit.blocks.26.block.self_attn.v,dit.blocks.27.block.self_attn.q,dit.blocks.27.block.self_attn.k,dit.blocks.27.block.self_attn.v,dit.blocks.28.block.self_attn.q,dit.blocks.28.block.self_attn.k,dit.blocks.28.block.self_attn.v,dit.blocks.29.block.self_attn.q,dit.blocks.29.block.self_attn.k,dit.blocks.29.block.self_attn.v,dit.blocks.30.block.self_attn.q,dit.blocks.30.block.self_attn.k,dit.blocks.30.block.self_attn.v,dit.blocks.31.block.self_attn.q,dit.blocks.31.block.self_attn.k,dit.blocks.31.block.self_attn.v,dit.blocks.32.block.self_attn.q,dit.blocks.32.block.self_attn.k,dit.blocks.32.block.self_attn.v,dit.blocks.33.block.self_attn.q,dit.blocks.33.block.self_attn.k,dit.blocks.33.block.self_attn.v,dit.blocks.34.block.self_attn.q,dit.blocks.34.block.self_attn.k,dit.blocks.34.block.self_attn.v,dit.blocks.35.block.self_attn.q,dit.blocks.35.block.self_attn.k,dit.blocks.35.block.self_attn.v,dit.blocks.36.block.self_attn.q,dit.blocks.36.block.self_attn.k,dit.blocks.36.block.self_attn.v,dit.blocks.37.block.self_attn.q,dit.blocks.37.block.self_attn.k,dit.blocks.37.block.self_attn.v,dit.blocks.38.block.self_attn.q,dit.blocks.38.block.self_attn.k,dit.blocks.38.block.self_attn.v,dit.blocks.39.block.self_attn.q,dit.blocks.39.block.self_attn.k,dit.blocks.39.block.self_attn.v" \
+  --extra_inputs "animate2_prompt_ref,animate2_reference_image,animate2_reference_video" \
+  --use_gradient_checkpointing \
+  --task "sft:train"
diff --git a/examples/wanvideo/model_training/lora/Wan2.2-Animate-2-14B-Distilled.sh b/examples/wanvideo/model_training/lora/Wan2.2-Animate-2-14B-Distilled.sh
new file mode 100644
index 000000000..481086f05
--- /dev/null
+++ b/examples/wanvideo/model_training/lora/Wan2.2-Animate-2-14B-Distilled.sh
@@ -0,0 +1,40 @@
+modelscope download --dataset DiffSynth-Studio/diffsynth_example_dataset --include "wanvideo/Wan2.2-Animate-2-14B-Distilled/*" --local_dir ./data/diffsynth_example_dataset
+
+accelerate launch examples/wanvideo/model_training/train.py \
+  --dataset_base_path data/diffsynth_example_dataset/wanvideo/Wan2.2-Animate-2-14B-Distilled \
+  --dataset_metadata_path data/diffsynth_example_dataset/wanvideo/Wan2.2-Animate-2-14B-Distilled/metadata.json \
+  --data_file_keys "video,animate2_reference_image,animate2_reference_video" \
+  --height 640 \
+  --width 352 \
+  --num_frames 81 \
+  --dataset_repeat 1 \
+  --model_id_with_origin_paths "Wan-AI/Wan2.2-Animate-2-14B:videomodel/Wan-AI/models_t5_umt5-xxl-enc-bf16.pth,Wan-AI/Wan2.2-Animate-2-14B:videomodel/Wan-AI/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth,Wan-AI/Wan2.1-T2V-14B:Wan2.1_VAE.pth" \
+  --learning_rate 1e-4 \
+  --num_epochs 5 \
+  --remove_prefix_in_ckpt "pipe.dit." \
+  --output_path "./models/train/Wan2.2-Animate-2-14B-Distilled_lora_splited_cache" \
+  --lora_base_model "dit" \
+  --lora_target_modules "self_attn.q,self_attn.k,self_attn.v" \
+  --lora_rank 32 \
+  --extra_inputs "animate2_prompt_ref,animate2_reference_image,animate2_reference_video" \
+  --use_gradient_checkpointing \
+  --task "sft:data_process"
+
+accelerate launch --config_file examples/wanvideo/model_training/full/accelerate_config_14B.yaml examples/wanvideo/model_training/train.py \
+  --dataset_base_path models/train/Wan2.2-Animate-2-14B-Distilled_lora_splited_cache \
+  --data_file_keys "video,animate2_reference_image,animate2_reference_video" \
+  --height 640 \
+  --width 352 \
+  --num_frames 81 \
+  --dataset_repeat 100 \
+  --model_id_with_origin_paths "Wan-AI/Wan2.2-Animate-2-14B:wan_animate_2/wan_animate_2_bf16_distillation.safetensors" \
+  --learning_rate 1e-4 \
+  --num_epochs 5 \
+  --remove_prefix_in_ckpt "pipe.dit." \
+  --output_path "./models/train/Wan2.2-Animate-2-14B-Distilled_lora" \
+  --lora_base_model "dit" \
+  --lora_target_modules "self_attn.q,self_attn.k,self_attn.v" \
+  --lora_rank 32 \
+  --extra_inputs "animate2_prompt_ref,animate2_reference_image,animate2_reference_video" \
+  --use_gradient_checkpointing \
+  --task "sft:train"
diff --git a/examples/wanvideo/model_training/lora/Wan2.2-Animate-2-14B.sh b/examples/wanvideo/model_training/lora/Wan2.2-Animate-2-14B.sh
new file mode 100644
index 000000000..2a812fcf2
--- /dev/null
+++ b/examples/wanvideo/model_training/lora/Wan2.2-Animate-2-14B.sh
@@ -0,0 +1,40 @@
+modelscope download --dataset DiffSynth-Studio/diffsynth_example_dataset --include "wanvideo/Wan2.2-Animate-2-14B/*" --local_dir ./data/diffsynth_example_dataset
+
+accelerate launch examples/wanvideo/model_training/train.py \
+  --dataset_base_path data/diffsynth_example_dataset/wanvideo/Wan2.2-Animate-2-14B \
+  --dataset_metadata_path data/diffsynth_example_dataset/wanvideo/Wan2.2-Animate-2-14B/metadata.json \
+  --data_file_keys "video,animate2_reference_image,animate2_reference_video" \
+  --height 640 \
+  --width 352 \
+  --num_frames 81 \
+  --dataset_repeat 1 \
+  --model_id_with_origin_paths "Wan-AI/Wan2.2-Animate-2-14B:videomodel/Wan-AI/models_t5_umt5-xxl-enc-bf16.pth,Wan-AI/Wan2.2-Animate-2-14B:videomodel/Wan-AI/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth,Wan-AI/Wan2.1-T2V-14B:Wan2.1_VAE.pth" \
+  --learning_rate 1e-4 \
+  --num_epochs 5 \
+  --remove_prefix_in_ckpt "pipe.dit." \
+  --output_path "./models/train/Wan2.2-Animate-2-14B_lora_splited_cache" \
+  --lora_base_model "dit" \
+  --lora_target_modules "self_attn.q,self_attn.k,self_attn.v" \
+  --lora_rank 32 \
+  --extra_inputs "animate2_prompt_ref,animate2_reference_image,animate2_reference_video" \
+  --use_gradient_checkpointing \
+  --task "sft:data_process"
+
+accelerate launch --config_file examples/wanvideo/model_training/full/accelerate_config_14B.yaml examples/wanvideo/model_training/train.py \
+  --dataset_base_path models/train/Wan2.2-Animate-2-14B_lora_splited_cache \
+  --data_file_keys "video,animate2_reference_image,animate2_reference_video" \
+  --height 640 \
+  --width 352 \
+  --num_frames 81 \
+  --dataset_repeat 100 \
+  --model_id_with_origin_paths "Wan-AI/Wan2.2-Animate-2-14B:wan_animate_2/wan_animate_2_bf16.safetensors" \
+  --learning_rate 1e-4 \
+  --num_epochs 5 \
+  --remove_prefix_in_ckpt "pipe.dit." \
+  --output_path "./models/train/Wan2.2-Animate-2-14B_lora" \
+  --lora_base_model "dit" \
+  --lora_target_modules "self_attn.q,self_attn.k,self_attn.v" \
+  --lora_rank 32 \
+  --extra_inputs "animate2_prompt_ref,animate2_reference_image,animate2_reference_video" \
+  --use_gradient_checkpointing \
+  --task "sft:train"
diff --git a/examples/wanvideo/model_training/train.py b/examples/wanvideo/model_training/train.py
index ce8fedcd5..45c941356 100644
--- a/examples/wanvideo/model_training/train.py
+++ b/examples/wanvideo/model_training/train.py
@@ -70,7 +70,7 @@ def parse_extra_inputs(self, data, extra_inputs, inputs_shared):
                 inputs_shared["input_image"] = data["video"][0]
             elif extra_input == "end_image":
                 inputs_shared["end_image"] = data["video"][-1]
-            elif extra_input == "reference_image" or extra_input == "vace_reference_image":
+            elif extra_input in ("reference_image", "vace_reference_image", "animate2_reference_image"):
                 inputs_shared[extra_input] = data[extra_input][0]
             else:
                 inputs_shared[extra_input] = data[extra_input]
diff --git a/examples/wanvideo/model_training/validate_full/Wan2.2-Animate-2-14B-Distilled.py b/examples/wanvideo/model_training/validate_full/Wan2.2-Animate-2-14B-Distilled.py
new file mode 100644
index 000000000..acb601ca7
--- /dev/null
+++ b/examples/wanvideo/model_training/validate_full/Wan2.2-Animate-2-14B-Distilled.py
@@ -0,0 +1,47 @@
+import torch
+from PIL import Image
+from diffsynth.utils.data import save_video, VideoData
+from diffsynth.core import load_state_dict
+from diffsynth.pipelines.wan_video import WanVideoPipeline, ModelConfig
+
+vram_config = {
+    "offload_dtype": torch.bfloat16,
+    "offload_device": "cpu",
+    "onload_dtype": torch.bfloat16,
+    "onload_device": "cuda",
+    "preparing_dtype": torch.bfloat16,
+    "preparing_device": "cuda",
+    "computation_dtype": torch.bfloat16,
+    "computation_device": "cuda",
+}
+
+pipe = WanVideoPipeline.from_pretrained(
+    torch_dtype=torch.bfloat16,
+    device="cuda",
+    model_configs=[
+        ModelConfig(model_id="Wan-AI/Wan2.2-Animate-2-14B", origin_file_pattern="wan_animate_2/wan_animate_2_bf16_distillation.safetensors", **vram_config),
+        ModelConfig(model_id="Wan-AI/Wan2.2-Animate-2-14B", origin_file_pattern="videomodel/Wan-AI/models_t5_umt5-xxl-enc-bf16.pth", **vram_config),
+        ModelConfig(model_id="Wan-AI/Wan2.2-Animate-2-14B", origin_file_pattern="videomodel/Wan-AI/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth", **vram_config),
+        ModelConfig(model_id="Wan-AI/Wan2.1-T2V-14B", origin_file_pattern="Wan2.1_VAE.pth", **vram_config),
+    ],
+    tokenizer_config=ModelConfig(model_id="Wan-AI/Wan2.2-Animate-2-14B", origin_file_pattern="videomodel/Wan-AI/umt5-xxl/"),
+)
+state_dict = load_state_dict("models/train/Wan2.2-Animate-2-14B-Distilled_full/epoch-1.safetensors")
+state_dict = {k.replace(".block.self_attn.", ".block.module.self_attn."): v for k, v in state_dict.items()}
+pipe.dit.load_state_dict(state_dict, strict=False)
+
+reference_image = Image.open("data/diffsynth_example_dataset/wanvideo/Wan2.2-Animate-2-14B-Distilled/refimage.jpg").convert("RGB")
+reference_video = VideoData("data/diffsynth_example_dataset/wanvideo/Wan2.2-Animate-2-14B-Distilled/refvideo.mp4").raw_data()
+num_frames = 81
+video = pipe(
+    prompt="人物外观描述：一名长黑发女性，穿着白色半透明蕾丝长袖上衣，衣身带有花卉刺绣，下身搭配白色百褶短裙和黑色腰带，脚穿米白色厚底运动鞋。 背景描述：背景为现代室内空间，墙面和柜体以浅灰色为主，后方设有两扇深色落地窗或玻璃门，顶部安装长条形灯具，中央有一块浅色长方形台面。",
+    negative_prompt="色调艳丽，过曝，静态，细节模糊不清，字幕，风格，作品，画作，画面，静止，整体发灰，最差质量，低质量，JPEG压缩残留，丑陋的，残缺的，多余的手指，画得不好的手部，画得不好的脸部，畸形的，毁容的，形态畸形的肢体，手指融合，静止不动的画面，杂乱的背景，三条腿，背景人很多，倒着走",
+    animate2_prompt_ref="视频中的人在做动作，背景静止",
+    animate2_reference_image=reference_image,
+    animate2_reference_video=reference_video[:num_frames],
+    animate2_offload_kv=True,
+    num_frames=num_frames, height=640, width=352,
+    num_inference_steps=40, cfg_scale=3.0,
+    seed=0, tiled=True,
+)
+save_video(video, "video_Wan2.2-Animate-2-14B-Distilled.mp4", fps=24, quality=5)
diff --git a/examples/wanvideo/model_training/validate_full/Wan2.2-Animate-2-14B.py b/examples/wanvideo/model_training/validate_full/Wan2.2-Animate-2-14B.py
new file mode 100644
index 000000000..2643b37c7
--- /dev/null
+++ b/examples/wanvideo/model_training/validate_full/Wan2.2-Animate-2-14B.py
@@ -0,0 +1,47 @@
+import torch
+from PIL import Image
+from diffsynth.utils.data import save_video, VideoData
+from diffsynth.core import load_state_dict
+from diffsynth.pipelines.wan_video import WanVideoPipeline, ModelConfig
+
+vram_config = {
+    "offload_dtype": torch.bfloat16,
+    "offload_device": "cpu",
+    "onload_dtype": torch.bfloat16,
+    "onload_device": "cuda",
+    "preparing_dtype": torch.bfloat16,
+    "preparing_device": "cuda",
+    "computation_dtype": torch.bfloat16,
+    "computation_device": "cuda",
+}
+
+pipe = WanVideoPipeline.from_pretrained(
+    torch_dtype=torch.bfloat16,
+    device="cuda",
+    model_configs=[
+        ModelConfig(model_id="Wan-AI/Wan2.2-Animate-2-14B", origin_file_pattern="wan_animate_2/wan_animate_2_bf16.safetensors", **vram_config),
+        ModelConfig(model_id="Wan-AI/Wan2.2-Animate-2-14B", origin_file_pattern="videomodel/Wan-AI/models_t5_umt5-xxl-enc-bf16.pth", **vram_config),
+        ModelConfig(model_id="Wan-AI/Wan2.2-Animate-2-14B", origin_file_pattern="videomodel/Wan-AI/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth", **vram_config),
+        ModelConfig(model_id="Wan-AI/Wan2.1-T2V-14B", origin_file_pattern="Wan2.1_VAE.pth", **vram_config),
+    ],
+    tokenizer_config=ModelConfig(model_id="Wan-AI/Wan2.2-Animate-2-14B", origin_file_pattern="videomodel/Wan-AI/umt5-xxl/"),
+)
+state_dict = load_state_dict("models/train/Wan2.2-Animate-2-14B_full/epoch-1.safetensors")
+state_dict = {k.replace(".block.self_attn.", ".block.module.self_attn."): v for k, v in state_dict.items()}
+pipe.dit.load_state_dict(state_dict, strict=False)
+
+reference_image = Image.open("data/diffsynth_example_dataset/wanvideo/Wan2.2-Animate-2-14B/refimage.jpg").convert("RGB")
+reference_video = VideoData("data/diffsynth_example_dataset/wanvideo/Wan2.2-Animate-2-14B/refvideo.mp4").raw_data()
+num_frames = 81
+video = pipe(
+    prompt="人物外观描述：一名长黑发女性，穿着白色半透明蕾丝长袖上衣，衣身带有花卉刺绣，下身搭配白色百褶短裙和黑色腰带，脚穿米白色厚底运动鞋。 背景描述：背景为现代室内空间，墙面和柜体以浅灰色为主，后方设有两扇深色落地窗或玻璃门，顶部安装长条形灯具，中央有一块浅色长方形台面。",
+    negative_prompt="色调艳丽，过曝，静态，细节模糊不清，字幕，风格，作品，画作，画面，静止，整体发灰，最差质量，低质量，JPEG压缩残留，丑陋的，残缺的，多余的手指，画得不好的手部，画得不好的脸部，畸形的，毁容的，形态畸形的肢体，手指融合，静止不动的画面，杂乱的背景，三条腿，背景人很多，倒着走",
+    animate2_prompt_ref="视频中的人在做动作，背景静止",
+    animate2_reference_image=reference_image,
+    animate2_reference_video=reference_video[:num_frames],
+    animate2_offload_kv=True,
+    num_frames=num_frames, height=640, width=352,
+    num_inference_steps=40, cfg_scale=3.0,
+    seed=0, tiled=True,
+)
+save_video(video, "video_Wan2.2-Animate-2-14B.mp4", fps=24, quality=5)
diff --git a/examples/wanvideo/model_training/validate_lora/Wan2.2-Animate-2-14B-Distilled.py b/examples/wanvideo/model_training/validate_lora/Wan2.2-Animate-2-14B-Distilled.py
new file mode 100644
index 000000000..44c2b6a71
--- /dev/null
+++ b/examples/wanvideo/model_training/validate_lora/Wan2.2-Animate-2-14B-Distilled.py
@@ -0,0 +1,44 @@
+import torch
+from PIL import Image
+from diffsynth.utils.data import save_video, VideoData
+from diffsynth.pipelines.wan_video import WanVideoPipeline, ModelConfig
+
+vram_config = {
+    "offload_dtype": torch.bfloat16,
+    "offload_device": "cpu",
+    "onload_dtype": torch.bfloat16,
+    "onload_device": "cuda",
+    "preparing_dtype": torch.bfloat16,
+    "preparing_device": "cuda",
+    "computation_dtype": torch.bfloat16,
+    "computation_device": "cuda",
+}
+
+pipe = WanVideoPipeline.from_pretrained(
+    torch_dtype=torch.bfloat16,
+    device="cuda",
+    model_configs=[
+        ModelConfig(model_id="Wan-AI/Wan2.2-Animate-2-14B", origin_file_pattern="wan_animate_2/wan_animate_2_bf16_distillation.safetensors", **vram_config),
+        ModelConfig(model_id="Wan-AI/Wan2.2-Animate-2-14B", origin_file_pattern="videomodel/Wan-AI/models_t5_umt5-xxl-enc-bf16.pth", **vram_config),
+        ModelConfig(model_id="Wan-AI/Wan2.2-Animate-2-14B", origin_file_pattern="videomodel/Wan-AI/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth", **vram_config),
+        ModelConfig(model_id="Wan-AI/Wan2.1-T2V-14B", origin_file_pattern="Wan2.1_VAE.pth", **vram_config),
+    ],
+    tokenizer_config=ModelConfig(model_id="Wan-AI/Wan2.2-Animate-2-14B", origin_file_pattern="videomodel/Wan-AI/umt5-xxl/"),
+)
+pipe.load_lora(pipe.dit, "models/train/Wan2.2-Animate-2-14B-Distilled_lora/epoch-4.safetensors", alpha=1)
+
+reference_image = Image.open("data/diffsynth_example_dataset/wanvideo/Wan2.2-Animate-2-14B-Distilled/refimage.jpg").convert("RGB")
+reference_video = VideoData("data/diffsynth_example_dataset/wanvideo/Wan2.2-Animate-2-14B-Distilled/refvideo.mp4").raw_data()
+num_frames = 81
+video = pipe(
+    prompt="人物外观描述：一名长黑发女性，穿着白色半透明蕾丝长袖上衣，衣身带有花卉刺绣，下身搭配白色百褶短裙和黑色腰带，脚穿米白色厚底运动鞋。 背景描述：背景为现代室内空间，墙面和柜体以浅灰色为主，后方设有两扇深色落地窗或玻璃门，顶部安装长条形灯具，中央有一块浅色长方形台面。",
+    negative_prompt="色调艳丽，过曝，静态，细节模糊不清，字幕，风格，作品，画作，画面，静止，整体发灰，最差质量，低质量，JPEG压缩残留，丑陋的，残缺的，多余的手指，画得不好的手部，画得不好的脸部，畸形的，毁容的，形态畸形的肢体，手指融合，静止不动的画面，杂乱的背景，三条腿，背景人很多，倒着走",
+    animate2_prompt_ref="视频中的人在做动作，背景静止",
+    animate2_reference_image=reference_image,
+    animate2_reference_video=reference_video[:num_frames],
+    animate2_offload_kv=True,
+    num_frames=num_frames, height=640, width=352,
+    num_inference_steps=40, cfg_scale=3.0,
+    seed=0, tiled=True,
+)
+save_video(video, "video_Wan2.2-Animate-2-14B-Distilled.mp4", fps=24, quality=5)
diff --git a/examples/wanvideo/model_training/validate_lora/Wan2.2-Animate-2-14B.py b/examples/wanvideo/model_training/validate_lora/Wan2.2-Animate-2-14B.py
new file mode 100644
index 000000000..6e3e2a4a6
--- /dev/null
+++ b/examples/wanvideo/model_training/validate_lora/Wan2.2-Animate-2-14B.py
@@ -0,0 +1,44 @@
+import torch
+from PIL import Image
+from diffsynth.utils.data import save_video, VideoData
+from diffsynth.pipelines.wan_video import WanVideoPipeline, ModelConfig
+
+vram_config = {
+    "offload_dtype": torch.bfloat16,
+    "offload_device": "cpu",
+    "onload_dtype": torch.bfloat16,
+    "onload_device": "cuda",
+    "preparing_dtype": torch.bfloat16,
+    "preparing_device": "cuda",
+    "computation_dtype": torch.bfloat16,
+    "computation_device": "cuda",
+}
+
+pipe = WanVideoPipeline.from_pretrained(
+    torch_dtype=torch.bfloat16,
+    device="cuda",
+    model_configs=[
+        ModelConfig(model_id="Wan-AI/Wan2.2-Animate-2-14B", origin_file_pattern="wan_animate_2/wan_animate_2_bf16.safetensors", **vram_config),
+        ModelConfig(model_id="Wan-AI/Wan2.2-Animate-2-14B", origin_file_pattern="videomodel/Wan-AI/models_t5_umt5-xxl-enc-bf16.pth", **vram_config),
+        ModelConfig(model_id="Wan-AI/Wan2.2-Animate-2-14B", origin_file_pattern="videomodel/Wan-AI/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth", **vram_config),
+        ModelConfig(model_id="Wan-AI/Wan2.1-T2V-14B", origin_file_pattern="Wan2.1_VAE.pth", **vram_config),
+    ],
+    tokenizer_config=ModelConfig(model_id="Wan-AI/Wan2.2-Animate-2-14B", origin_file_pattern="videomodel/Wan-AI/umt5-xxl/"),
+)
+pipe.load_lora(pipe.dit, "models/train/Wan2.2-Animate-2-14B_lora/epoch-4.safetensors", alpha=1)
+
+reference_image = Image.open("data/diffsynth_example_dataset/wanvideo/Wan2.2-Animate-2-14B/refimage.jpg").convert("RGB")
+reference_video = VideoData("data/diffsynth_example_dataset/wanvideo/Wan2.2-Animate-2-14B/refvideo.mp4").raw_data()
+num_frames = 81
+video = pipe(
+    prompt="人物外观描述：一名长黑发女性，穿着白色半透明蕾丝长袖上衣，衣身带有花卉刺绣，下身搭配白色百褶短裙和黑色腰带，脚穿米白色厚底运动鞋。 背景描述：背景为现代室内空间，墙面和柜体以浅灰色为主，后方设有两扇深色落地窗或玻璃门，顶部安装长条形灯具，中央有一块浅色长方形台面。",
+    negative_prompt="色调艳丽，过曝，静态，细节模糊不清，字幕，风格，作品，画作，画面，静止，整体发灰，最差质量，低质量，JPEG压缩残留，丑陋的，残缺的，多余的手指，画得不好的手部，画得不好的脸部，畸形的，毁容的，形态畸形的肢体，手指融合，静止不动的画面，杂乱的背景，三条腿，背景人很多，倒着走",
+    animate2_prompt_ref="视频中的人在做动作，背景静止",
+    animate2_reference_image=reference_image,
+    animate2_reference_video=reference_video[:num_frames],
+    animate2_offload_kv=True,
+    num_frames=num_frames, height=640, width=352,
+    num_inference_steps=40, cfg_scale=3.0,
+    seed=0, tiled=True,
+)
+save_video(video, "video_Wan2.2-Animate-2-14B.mp4", fps=24, quality=5)