ComfyUI Wan2.1 FLF2V & Wan2.1 Fun Native Support
We would like to share the latest update on Wan2.1 FLF2V native support in ComfyUI, as well as the Wan2.1 Fun series implementation earlier. These sets of models provide an extended ability for generating videos in ComfyUI.
ComfyUI Wan2.1 FLF2V Native Support
Wan FLF2V (First-Last Frame Video Generation) is an open-source 14B video model by the Alibaba Wan team. The model generates logically coherent and naturally flowing 720p high-definition videos.
Get Started
Update ComfyUI or ComfyUI Desktop to the latest
Download workflow below and drag into ComfyUI
The workflow will prompt you to download the models you need
Start generating!
Please note that we provide the FP16 repackaged model in the example for users to run on a local GPU. This quantized model trades a certain level of output quality for lower hardware requirements and faster speed.
Wan2.1 FLF2V 720P fp16 Workflow
Tip: A resolution of 720x1280 usually generates better results.
Example Outputs
Wan2.1 Fun Control & Wan2.1 Fun InP
Wan2.1 Fun is released by Alibaba, focusing on generating videos with more control:
Wan2.1 Fun Control: various control conditions, including Canny (line art), Depth, OpenPose (human posture), MLSD (geometric edges), and trajectory control
Wan2.1 Fun InP: first and last frame control
Get Started
Update ComfyUI or ComfyUI Desktop to the latest
Go to “Template” and select “Wan2.1 Fun Control” or “Wan2.1 Fun InP” workflows, or download workflows below
The workflow will prompt you to download the model
Start generating!
For manual file downloads, check our documentation for details.
Wan2.1 Fun Control Workflow
Example Outputs
Wan2.1 Fun InP Workflow
Example Outputs:
Please see our Documentation for more Community Resource and detailed instructions. Enjoy generating!