What changed in the model
Google has launched Gemini Omni 1.2 Flash with native video generation up to 15 seconds, compared with 10 seconds for Omni 1.1. Its developer announcement also describes multi-turn conversational editing and stronger following of image, motion and audio references.
The company lists a 720p rate of $0.07 per second, describing that as a 30% reduction from Omni 1.1. That is a generation rate for the stated configuration, rather than a complete budget for every production workflow or output resolution.
Editing becomes part of the same conversation
Google says users can change a background, lighting or aspect ratio through follow-up instructions while continuing work on a clip. Supported durations run from three to 15 seconds in one-second increments.
For a creator, the proposed benefit is preserving useful material while revising a particular detail. Franklin’s analysis is that this matters most when an almost-correct clip would otherwise require another complete generation. The announcement’s examples illustrate that behavior, but they do not establish an independent success rate for a user’s own characters, text or camera moves.
Availability and practical evaluation
Google describes general availability through the Gemini API and its business offering, with related rollouts across other products. Product-specific access and quotas still need to be checked separately; an API launch does not imply identical terms in every app.
The company acknowledges continuing work on fast sports action, rapid script changes and complex physics. A useful evaluation therefore includes the difficult part of the intended scene, rather than only a visually simple demo. Check reference retention, dialogue, timing and the cost of revisions together before estimating how many finished clips a budget will produce.