video_url on /v1/videos fails: edits go to /v1/video-router/generate
Sume's /v1/videos has frame_images and input_references but no edit field. Send video_url to /v1/video-router/generate with gemini-omni-flash-1.1.
You cannot edit a clip through POST /v1/videos. That route copies OpenRouter's body, which has frame_images and input_references but no source-video field, so Gemini Omni Flash 1.1's edit mode is a Video Router field. Send video_url to POST /v1/video-router/generate instead. Both routes create the same kind of job and read the same catalog ids.
Why the field is missing
/v1/videos infers the mode from which image field is present: frame_images means image-to-video, input_references means reference-to-video, and neither means text-to-video. A video_url entry inside input_references conditions a new generation, it does not edit. That is the difference between a reference clip and an edit source.
| Mode | /v1/videos | /v1/video-router/generate |
|---|---|---|
| Text-to-video | prompt | prompt |
| Image-to-video | frame_images | image_url (+ end_image_url) |
| Reference-to-video | input_references | reference_image_urls, reference_video_urls |
| Edit | Not available | video_url |
Mixing the two routes
The router uses Sume's { "data": ... } envelope and flat field names, while /v1/videos returns the bare OpenRouter job object. The Idempotency-Key header, job polling at /v1/jobs/{id}/status and the webhook signature work the same on both. Plan to use the router path only for edit, Recast and Motion Transfer, and keep /v1/videos for everything else.
- Edit request:
model,prompt,video_url, optionalresolution. - Do not send
aspect_ratioon an edit. - Poll with
GET /v1/jobs/{id}/status.
Sources
Related posts
More in Developers
- What balance do 100 video ads need? Sume reserves at submit
Sume reserves the credits of every accepted job at submit, so a 100-ad batch needs the whole total in the balance. 402 insufficient_credits otherwise.
- Why a video model is missing from your Sume model list
A documented video id can be absent from the model list: Veo needs Google Cloud, Tencent rows need the Tencent switch. Absent ids 404. Read 2026-10-13.
- X media upload: how many APPEND calls a Sume MP4 needs at 5 MB
X's chunked upload keeps each segment at or below 5 MB, with an 8 MB server limit. Divide your MP4's byte size by 5 MB, rounded up, to count APPEND calls.
- Year-in-review videos for 250 customers: bulk queues of 100, 100, 50
A Sume bulk run takes at most 100 items, so 250 year-in-review runs need three queues. Set concurrency 1 to 16, a spend cap per item and a new key per queue.
Written by Sume