Read a run's finished clips back off disk as a VIDEO list, ready for Video Assemble. Use it to join a film whose render was interrupted, without re-rendering…

Cloud AI in ComfyUI without an API key. Generate gpt-image-2 images and run GPT-5 text + vision through your existing ChatGPT/Codex login, plus Replicate models. 61 nodes: image gen, LLM, video, TTS, and a full product-variation pipeline.

comfyui-arkennemasis

folder

pattern

expected

Read a run's finished clips back off disk as a VIDEO list, ready for Video Assemble. Use it to join a film whose render was interrupted, without re-rendering anything.

Name	Type	Default	Description
folder	STRING		The run folder, e.g. ComfyUI/output/nemasis/story_001. Absolute, or relative to the ComfyUI output directory.
pattern	STRING	scene_(\d+)_\.mp4	Regex a filename must match. The first capture group, if there is one, is the scene number and decides the order.
expected	INT	00–999	How many clips there should be. 0 = accept whatever is there. Set it and a short run fails HERE, with a message naming the count, instead of quietly assembling an incomplete film.

Name	Type	Description
videos	VIDEO	—
report	STRING	—

arkennemasis Load Clips (finished clips from disk)

Inputs (3)

Outputs (2)