Overview
- Google rolled out Gemini Omni 1.2 Flash as a general‑availability product on Thursday, November 5, 2026, making the model available to developers and businesses through its official channels.
- The update raises native clip length to 15 seconds and adds multi‑turn conversational editing so users can refine the same clip through back‑and‑forth prompts instead of re‑generating.
- Omni 1.2 Flash improves frame‑to‑frame reference following and lip‑sync so faces, garments, packaging text, and brand colors remain consistent across camera moves and edits.
- Google cut the list price by 30% to $0.07 per second for 720p and recommends a draft‑then‑upscale workflow to save costs, while third‑party reporting shows API rates for 720p between $0.02 and $0.10 per second.
- The model is available across the Gemini API, Google AI Studio, Flow, Google Vids, the Gemini app, Google Cloud console, and partners like Runware and Segmind, and every output includes an imperceptible SynthID watermark though Google notes edge cases such as fast sports and complex physics still need work.