Gemini Omni 1.2 Flash adds 15-second video generation and conversational editing
Reported by Hacker News, Google DeepMind, Crypto Briefing and 6 more outlets9 sources ↓
Google says Gemini Omni 1.2 Flash is generally available through the Gemini API and Gemini for business, with native clips up to 15 seconds and editing through follow-up prompts. Google lists 720p video output at about $0.07 per second, 30% below Omni 1.1.
Key points
- Users can edit an existing clip with follow-up prompts instead of generating it again
- The paid Gemini API lists 720p video output at about $0.07 per second
- Access is rolling out across Google’s developer, business and consumer products
- Omni 1.1 API requests are scheduled to route to 1.2 starting December 3, 2026
Sources
- Hacker NewsRead ↗Gemini Omni 1.2 Flash
- Google DeepMindRead ↗Gemini Omni 1.2 Flash: Fast and cost-effective with longer clips
- Crypto BriefingRead ↗Google rolls out Gemini Omni 1.2 Flash with 15-second AI video generation
- Google CloudRead ↗Gemini Omni 1.2 Flash now GA: cost-effective video
- TipRanksRead ↗Google announces Gemini Omni 1.2 Flash
Details
The new model supports durations from 3 to 15 seconds, compared with a 10-second maximum for Omni 1.1. Google says users can revise clips through follow-up prompts, including changes to backgrounds, lighting and aspect ratio, while preserving other elements. The company also says the model better follows reference material and improves consistency, dialogue, lip-sync and camera movement.
Google lists 720p video output at about $0.07 per second, a 30% reduction from Omni 1.1. Unite.AI reports that a 15-second 720p clip costs about $1.04 in video-output charges; other input, text, retries and service costs may add to the bill. Google says users can draft at 360p or 720p and upscale selected clips to 1080p or 4K.
Google lists availability through the Gemini API, Google AI Studio, Google Cloud, Flow, Google Vids, Gemini for business, Google Ads and the Gemini app for Google AI Ultra subscribers. Google Cloud Provisioned Throughput is available for enterprise customers. Unite.AI reports that API requests to Omni 1.1 will route automatically to 1.2 starting December 3, 2026.
Google’s public API documentation, as reported by Kingy.ai and Unite.AI, limits first-to-last-frame interpolation to clips of 3 to 10 seconds and does not support uploaded audio inputs or audio references in the current API. Google DeepMind says development continues on fast sports action, rapid script changes and complex physics.
Written from coverage as of
Images
- Google DeepMind
- Unite.AI
- kingy.ai
- Tech Times
Stocks in this story
Change since open: 20 Nov to 21 Nov 2026
Last refresh