Google on Thursday, Nov. 5, launched Gemini Omni 1.2 Flash and made the video model generally available on the Gemini API and Google Cloud, according to Google DeepMind and Google Cloud posts.
The model can generate and edit videos natively up to 15 seconds long, up from 10 seconds with Omni 1.1, Google said. Duration can be set in one-second increments from 3 to 15 seconds. Users can draft at 360p or 720p, then upscale selected clips to 1080p or 4K. Pricing is $0.07 per second for 720p video, which Google said is 30% lower than Gemini Omni 1.1 Flash, with similar savings at other resolutions.
Google said Omni 1.2 Flash adds multi-turn conversational editing, so creators can refine the same clip turn by turn, such as swapping a product or background, instead of regenerating from scratch. It also claims stronger reference adherence for character, product or style images, short motion clips up to three seconds, and audio cues, plus more natural-sounding dialogue and clearer lip-sync. The Verge reported the 15-second length, conversational edits without full regeneration, and improved continuous camera moves and dialogue.
Outputs carry an imperceptible SynthID watermark, Google Cloud and the DeepMind consumer blog said. The DeepMind model card describes inputs of text, images, audio and video, and high-quality video with audio as output. It did not prominently detail C2PA provenance in the sections reviewed; SynthID is the watermark Google emphasized. Cloud notes that audio references are gated behind allowlisting.
Availability surfaces listed by Google include the Gemini API and Google AI Studio for developers, Gemini for business and Google Vids for enterprises and Workspace users, Flow for scene direction, the Gemini app for Google AI Ultra subscribers on a rollout, and Google Ads. Google said it is the first time Gemini Omni is generally available with Provisioned Throughput on Cloud.
Earlier Google frontier releases covered by AI Tech Daily include Gemini 4 Argon and Gemini 4 Carbon.
Sources
- Google DeepMind / Google Blog: Gemini Omni 1.2 Flash: Fast and cost-effective with longer clips, by Anish Nangia (November 5, 2026)
- Google Cloud Blog: Gemini Omni 1.2 Flash generally available, by John Nack (November 5, 2026)
- Google DeepMind: Gemini Omni 1.2 Flash model card (November 5, 2026)
- The Verge: Google’s Gemini Omni 1.2 Flash can generate clips up to 15 seconds long, by Emma Roth (November 5, 2026)
- AI Tech Daily: Google unveils Gemini 4 Argon with phased cyber-partner rollout (September 30, 2026)
- AI Tech Daily: Google unveils Gemini 4 Carbon with access limited to cyber defenders (October 28, 2026)