Google releases Gemini Omni 1.2 Flash with 15-second video generation, multi-turn editing and 30% lower price, generally available

Google made Gemini Omni 1.2 Flash generally available in the Gemini API, Google Cloud/Gemini for business, Google Vids and Flow. It is the first generally available release of the Omni line. Maximum clip length rises from 10 to 15 seconds, set in one-second steps from 3 to 15. Price falls 30% against Omni 1.1, to $0.07 per second at 720p. Other changes are multi-turn conversational editing, better adherence to reference images, motion clips and audio cues, improved lip-sync and fewer jump cuts. Output can be upscaled to 1080p or 4K, Provisioned Throughput is offered, and outputs carry a SynthID watermark.

Developers and businesses can now build on Omni with production-grade availability rather than preview terms. At $0.07 per second, a full 15-second 720p clip costs roughly $1.05 before upscaling, which makes iteration cheaper but does not make large volumes free. Editing a clip without regenerating it could cut wasted spend, if it works as described. Competing video-model vendors face a lower price reference point, though nothing here shows Google has pulled clearly ahead. The effect on video and marketing workers is plausible but unmeasured.

Claims of better consistency, reference adherence and lip-sync come only from Google, with no independent tests or side-by-side comparisons offered. "At scale" is marketing language, and the 30% cut is against Google's own earlier model, not rivals. Google itself concedes limits in fast sports action, rapid script changes and complex physics, so treat this as an incremental point release, not a step change. The price for 1080p and 4K upscaling is not stated in the extraction.

1. 01Independent side-by-side tests of Omni 1.2 Flash against rival video models on consistency, lip-sync and multi-turn edits, especially the known weak spots (sports, physics)
2. 02Real per-clip cost including 1080p/4K upscaling, rate limits and Provisioned Throughput pricing once developers publish bills, to check whether the 30% cut holds in practice
3. 03Competitor price responses or new releases from other video-model providers over the next 4 to 8 weeks, and any developer adoption signals such as API usage or product integrations