Google Launches Gemini Omni 1.2 Flash: 15-Second Video, 720p Price Cut 30%

What happened

The model went live on Nov 5 on the Gemini API and Gemini for business; Google says this is the first time Omni has reached general availability, with Provisioned Throughput for enterprises and an imperceptible SynthID watermark on outputs. Duration is the most visible change: native generation and editing run up to 15 seconds, half as long again as the 10 seconds in Omni 1.1, settable in one-second steps from 3 to 15 seconds, so a single clip can hold a complete ad spot or product explainer.

Editing changes with it: multi-turn conversational editing can swap the background, product, lighting, weather or aspect ratio turn by turn on the same 15-second clip, instead of regenerating the whole thing; creators can draft fast at 360p or 720p and upscale the keeper to 1080p or 4K. Reference following improves too: character, product or style images, motion clips of up to three seconds and audio cues can be combined to guide generation, keeping faces, garments, packaging typography and brand colors consistent across camera angles, with fewer jump cuts and steadier camera moves.

Voice and availability were announced together: multi-character dialogue gets clearer lip-sync, with each sampled voice routed to the right character; Flow, Google Vids, the Gemini app (rolling out to Google AI Ultra subscribers) and Google Ads all carry the model, and Google Vids users can generate 15-second clips at no additional cost with a Google account or Workspace subscription. Google adds that fast sports action, rapid script changes and complex physics are still being improved.

Key facts

1. Publisher and date: Google DeepMind, Nov 5, 2026; generally available on the Gemini API and Gemini for business
2. Duration: native generation and editing up to 15 seconds (10 seconds in Omni 1.1), settable from 3 to 15 seconds in one-second steps
3. Price: $0.07 per second for 720p, down 30% from Omni 1.1
4. Editing: multi-turn conversational editing that swaps background, product, lighting, weather or aspect ratio turn by turn; draft at 360p or 720p, upscale keepers to 1080p or 4K
5. Reference following: character, product or style images combined with motion clips of up to three seconds and audio cues
6. Voice: clearer lip-sync for multi-character dialogue, with each sampled voice routed to the right character
7. Camera: complex continuous camera moves with fewer jump cuts and steadier framing
8. Availability: Flow, Google Vids, the Gemini app (Google AI Ultra subscribers) and Google Ads; Provisioned Throughput for enterprises; SynthID watermark on outputs
9. Limits: fast sports action, rapid script changes and complex physics are still being improved