Google's first generally available Omni video model — 15-second clips, full-clip conversational editing, 4K upscaling, and a 30% price cut to $0.07/sec at 720p.
Last verified · 2026-11-05 · by Moe Ameen
Gemini Omni 1.2 Flash is the fast, cost-efficient tier of Google's Gemini Omni video family, made generally available on November 5, 2026. It is the first Omni model to leave preview and reach GA, backed by Google Cloud for production use. Where Omni 1.1 Flash was aimed at adding length and frame control, 1.2 is about maturity and economics: it is cheaper, its clips are longer, and its editing now spans the whole clip.
The concrete changes: a single generation runs up to 15 seconds, up from 10, with duration settable from 3 to 15 seconds in one-second steps and clip extension available to go longer. Conversational editing now works across the full 15-second clip — you can swap a product or background, change the lighting, weather, or aspect ratio, or adjust the camera angle turn by turn, instead of regenerating from scratch. You draft at 360p or 720p and upscale the final to 1080p or 4K. Google also cites stronger first-to-last-frame consistency, first/last keyframe interpolation, more natural speech with clearer lip-sync and multi-speaker conditioning, and fewer unwanted jump cuts on camera moves. Reference input accepts character, product, and style images, motion clips up to three seconds, and audio cues, and every output carries an invisible SynthID watermark.
The price is the headline. Google cut the 720p rate 30% to $0.07 per second, with similar savings at other resolutions. It is reachable through the Gemini API in Google AI Studio and the Google Cloud console — with Provisioned Throughput for production — and surfaces in Flow, Google Vids (15-second clips at no extra cost with a Google account or Workspace subscription), the Gemini app for Google AI Ultra subscribers, and Google Ads.
What has not changed is the shape of the thing. It is a video generation model, not a content workflow: no captions, no per-platform reframing, no brand-voice layer, no scheduler, and video only — no images, carousels, blogs, newsletters, or avatar talking heads. A cheaper, longer, editable 15-second clip is a better raw asset, but it is still a raw asset. For the earlier tiers, see [Gemini Omni 1.1 Flash](/ai-tools/gemini-omni-1-1-flash) and [Gemini Omni Flash](/ai-tools/gemini-omni-flash).
The underrated upgrade in 1.2 is not the 15 seconds — it is that conversational editing now spans the whole clip, so one generation can spawn a family of variants cheaply: the same scene in 9:16 and 1:1, with the background swapped or the lighting changed, each a quick chat turn away. That is exactly the raw material a multi-platform strategy wants. Kompozy is where those variants become native posts: feed the vertical cut into a TikTok/Reels/Shorts lane and the square or landscape cut into a LinkedIn/X lane, and Kompozy burns in branded captions, reframes cleanly per destination, and layers hook text through HyperFrames so the first second reads on silent autoplay. Then it schedules and fans the set across the eight social platforms plus your blog and email from one queue — so a single Omni 1.2 scene ships as five platform-tailored posts instead of five manual re-uploads.
The second half is breadth Omni does not have. Hand Kompozy the keeper clip and it fans the same idea into the formats the model can't make: a quote card, native text posts and a thread in your voice via your Persona Brief, a carousel, a blog draft, and a newsletter — plus Persona Shorts and HeyGen avatar video, and Clipped Shorts if you have long-form to cut from. At $0.07 a second you can afford to generate liberally; Kompozy is the part that turns that cheap volume into a scheduled, on-brand week of content across every platform.
It is the fast, cost-efficient tier of Google's Gemini Omni video family, made generally available on November 5, 2026 — the first Omni model at GA. It generates and conversationally edits short video up to 15 seconds and upscales to 1080p or 4K.
Google cut the price 30%: 720p video is $0.07 per second, with similar savings at other resolutions. You can draft cheaply at 360p or 720p and upscale the final to 1080p or 4K through the Gemini API.
It is generally available rather than preview, raises a single clip to 15 seconds (from 10), extends conversational editing across the whole clip, improves consistency and lip-sync, adds Provisioned Throughput for production, and cuts the 720p price 30% to $0.07 per second.
No. It generates and edits a clip but has no captioning, per-platform reframing, scheduling, or posting. Bring the export into Kompozy to caption, reframe, schedule, and publish it across TikTok, Reels, YouTube Shorts, X, LinkedIn, and more — and to fan the same clip into a quote card, text posts, and a thread in your voice.