Skip to content

Google releases Gemini Omni 1.2 Flash video model

Google AI

The model is generally available via the Gemini API, Google Cloud, and Gemini for business at $0.07 per second for 720p video, a 30% price cut.

Key points

  • Generates native video clips from 3 to 15 seconds in one-second increments, up from a 10-second limit in version 1.1
  • Supports multi-turn conversational editing to change backgrounds, lighting, and aspect ratios without full regeneration
  • Accepts reference images, style cues, audio prompts, and up to three-second motion clips to maintain consistency
  • Allows drafting at 360p or 720p before upscaling final outputs to 1080p or 4K
  • Generates multi-speaker dialogue with synchronized lip movements and tags outputs with SynthID watermarks

Why it matters

Developers and creators can produce longer video clips with iterative editing rather than paying to re-render full scenes.

Sources