Browse / comfyui

First frame + last frame → video — ComfyUI workflow

A clip that starts on one picture and ends on another.

By Plops · Updated

First frame + last frame → video

A clip that starts on one picture and ends on another.

What it takes

  • A first frame
  • A last frame
  • A prompt describing what happens in between

What it gives back

  • An MP4, played inline and downloadable

What actually runs

  • Wan 2.1 first-last-frame-to-video, 720p weights in fp8.
  • Both frames are encoded through CLIP vision and handed to the model as the two ends of the shot.
  • Frames are decoded and muxed to H.264.

Worth knowing. Keep the two frames in the same framing. The model interpolates a shot, not a cut — wildly different compositions produce a smear.

Running it

The .json attached to this listing is the graph itself — the one the Forge runs, not a tidied-up copy. Values in {{double braces}} are the placeholders filled in per job; replace them with real values and it runs in any ComfyUI new enough for the nodes it uses.

Full setup — the model manifest, the directories, the traps — is at /workflows/video.

Licence and credit

The graph is ours and you may use it for anything, including commercially. The models it loads are not ours: each carries its own licence from whoever trained it, and that is the one that governs what you may do with the output.

License: MIT (the graph); models keep their own