Skip to main content
Video Split

Video Split · screenshot coming soon

​
What this node does

Video Split splits a clip into a before part and an after part at the timestamp split_s. It’s a real re-encode on the server (PyAV / ffmpeg): on ▶ Run it trims the clip twice (0 → split, and split → end) and saves both as project snapshots. Input is COMFYTV_VIDEO; it has two COMFYTV_VIDEO outputs. The split point must fall inside the clip — roughly between 0.05s and (duration − 0.05s) — or the run errors out.

​
When to use it

  • Break a long take into two shots so you can treat each one separately (retime, color, etc.).
  • Isolate a section, then feed one half into Video Concat with other material.
  • Drop a mid-clip transition point.

​
Parameters

​
split_s

The cut point, in seconds. Range 0–3600, step 0.01, default 0.0. Must land strictly inside the clip; a value at or beyond the ends is rejected. Everything before this time goes to video_a, everything from it to the end goes to video_b.

​
Outputs

Both parts are persisted as snapshots; video_b is emitted as the “picked” side of the same run.

​
Tips

  • The default 0.0 is not a valid split — set a real timestamp inside the clip.
  • Duration is read from the clip automatically; the split must sit at least ~0.05s away from either end.

​
Bridge note

COMFYTV_VIDEO is a project snapshot reference, not a native ComfyUI tensor. To move between ComfyTV and native ComfyUI nodes, insert a Bridge (ComfyTV/Bridge). See https://github.com/jtydhr88/ComfyTV/blob/main/docs/bridges.md.
  • Video Clip — trim to an arbitrary in/out range instead of a single cut.
  • Video Concat — rejoin the two halves (with or without other clips).