Seedance2.5: Video Editing
Edit existing video clips with Seedance 2.5 by describing changes in natural language, enabling local object removal, background swaps, scene replacements, and style transfers while preserving the original composition and motion. Feed in 1 video and receive 1 edited video output with temporally consistent results across frames. Ideal for e-commerce background replacement, stylized remakes of footage, and ad re-editing with targeted retouching.
概要
Seedance2.5: Video Editing is a text-driven video-to-video editor built in ComfyUI. It loads your source clip with LoadVideo, optionally trims it with Video Slice, and sends the frames to ByteDance2ReferenceNode running Seedance 2.5 in video_editing mode. You describe the change in plain language—such as “replace the background with a minimalist white studio” or “remove the product label and keep everything else the same”—and the model returns a single, temporally consistent edited clip.
入力するもの
- ワークフローを読み込み中
ComfyUI で実行できます。 このワークフローはこのページでは実行できません。すべてが読み込まれた状態でComfy Cloudで開くか、ご自身のComfyUI用にダウンロードしてください。
Cloudで試すよくある質問
It switches the model into edit mode so the output follows your source video’s composition, motion, aspect, and timing. Your text instruction is applied as a guided change rather than generating a completely new video.
Yes. In video_editing mode, the workflow inherits the input video’s aspect and duration, so timing and framing remain intact while the edit is applied.
Keep instructions concise and unambiguous, work on shorter slices with Video Slice (avoid hard cuts within the same run), and iterate in small steps. If the node exposes quality/consistency controls, prefer defaults or moderate values rather than extreme settings.
SaveVideo typically outputs a silent clip in ComfyUI. If you need audio, export the edited video and reattach the original soundtrack in a video editor.
Because ByteDance2ReferenceNode uses an API, processing time depends on clip length, resolution, and network speed. For faster, more stable edits, start with 5–15 second segments at 720p–1080p. If you encounter errors or long waits, reduce duration or resolution.