FLUX video edit API: 720p downscale, and video_url edits on Sume
BFL's edit-a-video endpoint downscales sources above 720p. Sume lists no FLUX video; its edit path is video_url with gemini-omni-flash-1.1 only.

BFL's "Edit a video" endpoint sets duration, resolution, aspect ratio and audio from your source clip and downscales sources larger than 720p to 720p. Sume has no FLUX video model; its video-to-video edit is the video_url field with gemini-omni-flash-1.1, the only model that accepts it.
BFL's text is from its docs index (the page is undated); Sume's from the Video Router docs, read 2026-09-30.
What does the BFL endpoint do?
The index describes it as transforming a source clip with an edit instruction. The server sets duration, resolution, aspect ratio and audio from the source, and sources above 720p are downscaled to 720p. It does not accept FLUX 3 generation controls such as mode, version, seed or generate_audio.
How does editing work on Sume?
| Topic | Rule |
|---|---|
| Model | gemini-omni-flash-1.1 only |
| Request | video_url plus a prompt describing the edit |
| Resolution | Optional; default 720p |
| aspect_ratio | Rejected |
| Duration | Not sent to the provider; only a reserve-estimate hint |
| Combining | Cannot be combined with image_url, end_image_url or reference_*_urls |
What does that mean for source clips?
Gemini Omni Flash 1.1 is documented at 3 to 10 seconds for generation. The docs state no source-length limit and no 720p downscale rule, so test a short clip first. The clip is the edit source, not a reference: to condition a fresh generation on a clip instead, the schema points to reference_video_urls.
What does an edit request look like?
POST to /v1/video-router/generate with an Idempotency-Key, model: "gemini-omni-flash-1.1", your prompt, video_url and mode: "async". Prompt wording tips are in Gemini Omni video edit prompts.
Sources
Related posts
More in Models
- gemini-2.5-flash-image shutdown on October 2, 2026
Google lists gemini-2.5-flash-image for shutdown on October 2, 2026. Sume lists Nano Banana 2 and Pro under google/ ids to move image requests to.
- Gemini 3.5 Transcribe API limits: 1 hour vs Sume STT's 10 minutes
Gemini 3.5 Transcribe takes up to 1 hour per request, 30 minutes with diarization or word timestamps. Sume STT takes 600 seconds; split longer audio.
- Gemini API speaker diarization: 8 speakers; Sume STT has no labels
Gemini 3.5 Transcribe diarizes up to 8 speakers, 3+ experimental, within 30 minutes per request. Sume STT returns word timings but no speaker labels.
- Gemini Omni video editing: the EEA, UK and under-10-second rules
Google says editing or extending uploaded videos with Omni is unavailable in the EEA, Switzerland and the UK, and uploads must be under 10 seconds.
Written by Sume