Gemini Omni in Google Vids: where editing your own video is blocked

Yes, Google Vids can edit a video you uploaded with Gemini Omni, but not everywhere. Google's Workspace Updates post says editing non-AI videos with Omni is unavailable in the European Economic Area, Switzerland, the United Kingdom, Texas and Illinois. Generating a new clip is a separate feature, and the post lists no regional block on it.

This page reads Google's Generate higher quality AI video clips and edit any video with Gemini Omni in Vids announcement, read on 2026-10-02. For the Sume side it uses the Video Router docs. Nothing here is legal advice, and Google's page is the only source for what Google allows.

What does Omni in Vids actually do?

Google describes two things. The first is generation: higher quality video with improved text rendering, physics and realism, from Google's latest Omni Flash model. The second is editing: you change an existing video with text instructions, and the post's examples are adjusting color grading, applying visual styles and removing background noise.

The post does not state a clip length limit or any other numeric usage limit, so do not plan around a number for Vids. The numbers Google publishes for Omni live on the Gemini API pages, such as the 10-second input cap for editing, and those belong to the API, not to Vids.

Which plans and regions are covered?

The announcement lists the editions that get the feature. Region limits apply only to editing footage that was not made by AI.

What if my team sits in a blocked region?

The block is a rule about Google's product. If your team is in Texas or the EEA and needs to restyle a clip you filmed, Vids will not do it, and no setting changes that. Your options are to edit with ordinary tools, or to use a different service for that step.

Sume lists Gemini Omni Flash 1.1 in its video catalog, and its edit mode takes one source clip in video_url plus a prompt. The Video Router page describes that route without any regional carve-out, but it also does not say the question is closed: whether you may upload a given clip is a matter for your own terms, consent and content rules. Treat the lack of a stated block as an absence of documentation, not as permission.

How do I run the same edit on Sume?

Send the clip's URL and an instruction that ends by pinning everything else. The edit mode is video_to_video: resolution is optional and defaults to 720p, and aspect_ratio and duration are not accepted because the output follows the source. Native audio is always on for this model, so you cannot ask for a silent edit. The request below is the documented shape.

Submit it as an async job, keep the job id, and read the status route when you come back. Do not resubmit a paid request because your own process timed out; the Jobs and results page says the same.

curl -X POST https://api.sume.com/v1/video-router/generate \
  -H "Authorization: Bearer $SUME_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: vids-style-edit-001" \
  -d '{
    "model": "gemini-omni-flash-1.1",
    "prompt": "Apply a warm film color grade. Keep everything else the same.",
    "video_url": "https://example.com/clip.mp4",
    "resolution": "720p",
    "mode": "async"
  }'

What does Sume not do here?

Pick Vids when you are inside Workspace in an allowed region and want no code. Pick an API job when you need a job id, a retry story and a record of what ran.

• It is not a Vids add-on. There is no Workspace integration; you call an API or an agent.
• It does not remove background noise as a separate step. The edit is one prompted pass of the video model, so a result can change more than you asked.
• The video_url source cannot be combined with image_url, end_image_url or reference lists.
• Price is the provider list rate times 1.25 per output second, by resolution. Read live numbers from GET /v1/video-router/models, not from a blog post.

How do I decide between Vids and an API call?

Start from who needs to run the edit. Vids sits inside Workspace, has no admin controls for this feature and is tied to the editions listed above, so it suits a team that already lives in Google's apps and works in a permitted region. An API call suits a process: a batch of clips, a nightly job, an agent that edits on request.

Three practical checks keep either path honest. First, confirm which edition your account is on, because the announcement lists them by name and a plan not on the list does not get the feature. Second, confirm where the people who will click the button are, since the region rule follows Google's product, not your company's headquarters. Third, test on one clip and compare it frame by frame to the original, because a prompted edit can change things you did not mention.

Whatever you pick, keep the original file. Neither Google's post nor Sume's docs promise that an edit is reversible, and a Sume edit is a new output that leaves the source alone, which is why the original is still yours to return to.