Skip to main content

Frame Extractor

The Frame Extractor node pulls a single still image out of a video. Choose the first or last frame, and the node outputs it as an image other nodes can consume.

Inputs

Outputs

How it works

  1. Receives a video from the input handle
  2. Grabs the chosen frame — the opening frame or the closing frame
  3. Outputs that frame as an image to the next node

Configuration options

  • Which Frame — toggle between First Frame and Last Frame. Defaults to First Frame.
No engine selection needed — this is a utility node. It runs in the background (auto-run): because the only choice is made here at design time, the node extracts the frame silently during the run with no manual step. Place a Review node after it if you want to eyeball the still before the rest of the flow continues.

Credit cost

Free — 0 credits.

Relationship to sequential Split

Split (For Each) in sequential mode already does an automatic last-frame handoff between iterations — Image Generation grabs the previous shot’s final frame on its own, with no node to place. That’s the right tool when the number of shots is dynamic (e.g. a script parsed into a variable number of scenes). The Frame Extractor is the explicit counterpart for linear, fixed-length flows where you’ve hand-wired a known number of shots and want the handoff visible on the canvas:
The two are not interchangeable: Frame Extractor is a single data node, not an iteration construct, and one placed inside a loop body can’t see the previous iteration’s video. Use sequential Split for dynamic shot counts; use Frame Extractor for explicit, fixed-length chains (or any one-off “grab a still from this video” need).

Tips

  • Feed the last frame into an Image-to-Video node to continue a shot from where the previous clip ended, keeping motion and composition consistent.
  • Use the first frame to generate a thumbnail or a starting reference image for a new variation.
  • Pair it with Image Generation to restyle a single moment from a clip.

Example use cases

  • Extending a video by continuing from its final frame
  • Creating a cover image / thumbnail from a generated clip
  • Capturing a reference frame to drive a continuity-consistent next shot