Gemini Omni 1.1 Flash has no shutdown date yet: how to pin and watch
Google lists gemini-omni-1.1-flash with no shutdown date announced, while Veo 3.1 previews end October 22. Pin the id, and watch the deprecations page.
Google's deprecations page lists gemini-omni-1.1-flash as released August 27, 2026 with "No shutdown date announced." That is a statement about today, not a promise. The three Veo 3.1 preview ids, by contrast, all end on October 22, 2026, and Google names the Omni model as their replacement. For a pinned pipeline, the practical rule is to pin the exact id and read the deprecations page on a schedule.
How long Google's video ids have lasted
Dates are from the Gemini API deprecations page. The durations are my own subtraction from those dates.
| Model id | Released | Shutdown | Time in service |
|---|---|---|---|
| veo-3.1-generate-preview | October 15, 2025 | October 22, 2026 | 12 months and 7 days |
| veo-3.1-fast-generate-preview | October 15, 2025 | October 22, 2026 | 12 months and 7 days |
| veo-3.1-lite-generate-preview | March 31, 2026 | October 22, 2026 | 6 months and 22 days |
| veo-3.0-generate-001, veo-3.0-fast-generate-001, veo-2.0-generate-001 | not stated here | June 30, 2026 | already shut down |
| gemini-omni-1.1-flash | August 27, 2026 | No shutdown date announced | ongoing |
What the changelog adds
The Gemini API changelog had no October 2026 entries when I read it, and its last two relevant entries were the August 27 general availability of Omni and a September 1 note on agentic video understanding. Nothing on that page changes the Omni status.
Pinning on Sume
On Sume, the catalog id is gemini-omni-flash-1.1, and the default of sume/auto resolves to it for text, first-frame, end-frame and reference requests. The Sume docs also say that resolution is a pure function of the request plus the catalog version. So a request to sume/auto is stable inside one catalog version, but a later catalog version can change what the default resolves to.
If your product needs one model for a long time, send the named id. If you want Sume to move you when the catalog moves, send sume/auto. Either way, the poll response echoes what you sent, so log your own request.
| Choice | What you send | What stays fixed |
|---|---|---|
| Pinned | model: "gemini-omni-flash-1.1" | The model family, until the id leaves the catalog |
| Default | model: "sume/auto" | Route and price inside one catalog version only |
A watch routine
Put these four items in the repo, not in someone's memory.
- Store the model id in one config value, not in prompts or scripts.
- Check the deprecations page when you cut a release, and add the check to your release checklist.
- Run a 3 to 10 second canary generation on each deploy so a retired id fails in a test and not in production.
- Read
GET /v1/videos/modelsfor the live catalog; a missing id returns404 model_not_found.
Sources
Related posts
More in Developers
- Gemini Omni reference clips: 3 videos, 3 seconds each, VIDEO_REF_0
On Sume, Gemini Omni Flash reference-to-video takes up to 10 images and 3 clips of at most 3 s each, addressed as IMAGE_REF_0 and VIDEO_REF_0 in the prompt.
- Gemini Omni Flash edit on Sume: the 8-second reserve hint
A Gemini Omni edit has no duration input, so Sume reserves against a hint: 8 seconds by default, up to 30 if you send one. At 720p, that is $1.00 held.
- Gemini Omni Interactions API versus Sume /v1/videos: field mapping
Google returns Omni video as base64 in an Interactions response. Sume returns 202 with a polling_url and a content URL. See the mapping and a bytes helper.
- Gemini Omni prompt template: five parts in a Python builder
Google's Omni guide names five prompt parts: framing, style, lighting, location, action. Build them in Python and send the prompt to Sume's /v1/videos.
Written by Sume