New Gemini model family launched at I/O 2026: image + audio + video + text in → video out. Distinct from the LLM-tier Gemini 3.5 Flash — positioned as a "create anything from any input" world-simulating generator with grounding in real-world physics.

Initial release is Gemini Omni Flash: 10-second video cap, SynthID-watermarked. Available globally to Gemini app subscribers (AI Plus / Pro / Ultra) and Google Flow; free tier via YouTube Shorts and YouTube Create. Direct competitive answer to Sora / Veo separation.

Gemini Omni 1.2 Flash (November 5, 2026) is generally available on the Gemini API and Gemini for business as gemini-omni-1.2-flash. It raises native clip length to 15 seconds, "up from 10 seconds with Omni 1.1", settable in one-second steps from 3 to 15, and adds multi-turn conversational editing and stronger reference following. Google cut the price 30% from Omni 1.1, to $0.07 per second of 720p video; drafts can be made at 360p or 720p and upscaled to 1080p or 4K. The post gives no benchmark numbers and no architecture details. (This index has no separate record of the Omni 1.1 release.)

Model Details

Variants

Name Parameters Notes
Gemini Omni Flash — First public release; 10s cap; SynthID-watermarked; in Gemini app + YouTube Shorts/Create
Gemini Omni 1.2 Flash — 2026-11-05; GA on Gemini API; 15s clips (up from 10s in Omni 1.1); multi-turn editing; $0.07/s at 720p, 30% below Omni 1.1
generationmultimodalvideo

Related