Gemini Omni 1.2 Flash

Not yet independently verified. Not yet independently verified. Capabilities and prices are from Google's own changelog, model page and pricing page, and the release date is also shown on OpenRouter; we found no independent test. We will update this when it can be confirmed, and remove this note.

Google made Gemini Omni 1.2 Flash generally available in the Gemini API on 5 November 2026, under the model code gemini-omni-1.2-flash. It takes text, images and video up to 15 seconds long, and returns video with native audio as well as text.

Fifteen-second clips you can edit by talking to them

A single generation runs from 3 to 15 seconds. Editing is conversational, so a clip can be changed turn by turn, and each turn can extend it by 3 to 10 seconds up to a total of 30 seconds. Interpolating between a first and last frame covers 3 to 10 seconds. Up to six reference images keep a character or subject consistent across shots. Output is 24 frames a second at 360p, 720p, 1080p or 4K.

What a clip costs at each resolution

The bill is in tokens, which makes the arithmetic less obvious than a per-second price. Google counts 5,792 output tokens per second of 720p video, which comes to about $0.07 a second, about $0.70 for 10 seconds and about $1.04 for 15. It gives $0.023 a second at 360p, $0.104 at 1080p with upsampling and $0.209 at 4K with upsampling. There is no free tier. Google also doubled paid-tier rate limits with this release.

Requests to Gemini Omni 1.1 Flash route to this model automatically from 3 December 2026, at 1.2 rates.

Key facts

Maker: Google Released: 5 November 2026 (generally available) Model code: gemini-omni-1.2-flash Input limit: 131,072 tokens Output limit: 65,536 tokens Price per million tokens: $1.50 input, $9 text output, $12 video output Availability: Gemini API, paid tier