On November 5, 2026, Google announced Gemini Omni 1.2 Flash as generally available: the paid Gemini API tier and Google Cloud went live the same day, the official blog published that day, and the Google Cloud blog followed the next day. It's the second GA generation of the Gemini Omni family — under six months from the family's May 19 I/O announcement to 1.2 GA.
On May 19, 2026, Google I/O announced the Gemini Omni family ("any-to-any", starting with video, image and audio output to follow), live the same day in the Gemini app, Flow and YouTube Shorts. On June 30, Gemini Omni Flash entered public preview on the Gemini API as gemini-omni-flash-preview. On August 27, Gemini Omni 1.1 Flash went GA, adding scene extension (40s cumulative), first/last-frame control, a 360p draft tier and 1080p/4K upscaled output; the preview retired Sep 30. On Oct 22, Veo 3.1's Gemini API preview endpoints retired, with Omni as the official migration target — by the time 1.2 launched, Omni was already the only video-model line on the Gemini API.
On Nov 5: the paid Gemini API tier and Google Cloud went generally available; Google Vids switched to the 1.2 engine and opened free quotas the same day (the official blog says any Google account or Workspace subscriber gets up to 3 free 15-second videos per month); Flow retired Veo Extend the same day. The Google Cloud blog followed the next day (by John Nack), confirming the Cloud-side GA and listing third-party customer quotes (Artlist, FLORA, invideo, Krea, Picsart, Runway, Photoroom, QuickFrame AI). Third-party platforms moved fast: Higgsfield, Segmind and OpenRouter listed 1.2 the same day or the next; TeamoRouter hadn't added it as of Nov 8.
The official blog makes three claims. First, duration: generation and editing expanded from 10s to 15s (whole-second steps, default 10s), extension 3–10s per turn, 30s total. Second, price: video output fell from $17.50 to $12.00/1M tokens — Google says 30% cheaper (exactly 31.4%); 720p ~$0.07/s, 15s ~$1.04. Third, quality claims: faster generation, stronger reference adherence, more consistent characters/objects/scenes, more natural dialogue, steadier camera motion, fewer jump cuts — but these are vendor claims; Google published no benchmark scores of its own, and AccessAllGPT explicitly says it didn't reproduce them.
An early brief said 1.2 was GA on "the paid Gemini API tier + Gemini for business" — corrected: the GA covers the Gemini API and Google Cloud; the Cloud blog says in-chat video generation inside Gemini for business is "Coming soon", not live yet. The Gemini app's 15s video is also still rolling out: the official blog says "rolling out to Google AI Ultra subscribers", Plus/Pro users are still at 10s, and as of Nov 13 Google hasn't announced the rollout completed.
On launch day, the AA video board hadn't listed 1.2 yet; scores appeared around Nov 10: text-to-video 1154 (#6, tied with Seedance 2.5), image-to-video 1086 (#14), editing 1164 (#2). The first tests agree: instruction-following is strong (Segmind calls it "the most instruction-following video model tested"), chart animation and single-point edits on real footage are strongest; but a 360p draft and the same-prompt 720p are completely different scenes — 360p cannot be a cheap preview. What to watch next: the Dec 3 auto-routing of 1.1, Flow's Dec 5 retirement of 1.1, news of Omni Pro (teased by Google, no date), and how the AA scores move with votes. This site will keep tracking it.
Reference: Google — Introducing Gemini Omni 1.2 Flash (rewritten and organized by ModelBrief; facts per the official source)