Google has made Gemini Omni 1.2 Flash generally available, its latest AI model for creating and editing short video with sound. Google announced the launch on 5 November, describing it as faster, cheaper and more consistent than the version it replaces.

The headline change is length. Clips can now run for 15 seconds, up from 10 seconds with Omni 1.1 Flash, and the duration can be set in one-second steps from three to 15 seconds. Google says that is enough for a complete product explainer or a short social advertisement in a single generation, without stitching shorter clips together.

Every clip comes with native audio, including dialogue with clearer lip-sync and distinct voices when more than one person is speaking. The model can also be steered with reference material, such as a character or product photograph, a style image, or a motion clip of up to three seconds, and Google says faces, packaging and brand colours now hold together better from the first frame to the last.

Editing works in conversation. Instead of starting again when a detail is wrong, a user can type a follow-up instruction, for example to change the background, lighting or weather, and the model edits the same clip. Drafts can be made at 360p or 720p and then upscaled to 1080p or 4K, and every output carries Google's invisible SynthID watermark.

Pricing has fallen by 30% from the previous version, to $0.07 per second for 720p video, so a full 15-second clip costs about $1.05 through the paid API tier. From 3 December, requests sent to Omni 1.1 Flash will be routed to 1.2 automatically. Google is candid about the limits: fast sports action, rapid script changes and complex physics are still being improved.