Wan 2.2 FLF2V (First Last Frame to Video) is an upgraded open-source video generation model that produces smooth, coherent videos between two input frames. Built with native ComfyUI support and developed by the ComfyUI team, it improves motion consistency, detail retention, and overall video quality compared to Wan 2.1.

Simply input a first and last image and the model creates a 720p interpolated video — ideal for storytelling, design transitions, and animation pipelines.

✨ Key Features

Frame-to-Frame Interpolation: Define start and end frames; the model fills in all motion between them.
Wan 2.2 Backbone: Upgraded model with better motion consistency, detail retention, and temporal coherence.
720p Output: High-fidelity video generation at cinematic resolution.
Native ComfyUI: Built-in support — no external dependencies or complex wiring needed.

🚀 Getting Started

Upload your first frame and last frame images.
Configure resolution and frame count as needed.
Generate — Wan 2.2 FLF2V interpolates all frames and outputs a smooth video.

Click the "Run Directly" link above to bypass local setup and test this workflow immediately in your browser.