Google on November 5, 2026 made Gemini Omni 1.2 Flash generally available across the Gemini API and Gemini for business. The Google DeepMind product post and the matching Google Cloud blog describe native video clips up to 15 seconds, multi-turn conversational editing, stronger reference following, and a 30% price cut versus Omni 1.1, to $0.07 per second at 720p.

The company says this is the first time Gemini Omni is generally available for production use, with Provisioned Throughput on Google Cloud and an imperceptible SynthID watermark on outputs.

Key Facts

Item Detail
Announced November 5, 2026 (Google / DeepMind blogs)
Status Generally available on Gemini API and Google Cloud
Max clip length 15 seconds (was 10 on Omni 1.1)
Duration control 3–15 seconds in 1-second steps
API price (720p) $0.07 / second (30% below Omni 1.1); similar savings at other resolutions per Cloud blog
Draft / upscale Draft at 360p or 720p; upscale selected clips to 1080p or 4K
Editing Multi-turn conversational edits on the same clip (swap product, background, lighting, aspect ratio)
References Character, product, or style images; motion clips up to 3 seconds; audio cues (Cloud blog: audio references are allowlist-gated)
Watermark Imperceptible SynthID
Consumer surfaces Flow; Google Vids (up to three 15-second clips per month at no cost with a Google account or Workspace); Gemini app rolling out to Google AI Ultra; Google Ads

Google's own post flags remaining weak spots: fast sports action, rapid script changes, and complex physics are still being improved.

Background

Omni 1.1 topped out at 10-second clips. Omni 1.2 Flash stretches that to a full 15-second generation or edit, which matches a common short-form ad length and removes the need to stitch two 10-second takes for many social spots.

A separate Workspace post says Google Vids now runs on Omni 1.2 Flash and lets anyone with a Google account or Workspace subscription generate up to three 15-second videos per month at no cost. The DeepMind post adds Flow for scene direction and multi-turn edits, Gemini-app access for Google AI Ultra subscribers as it rolls out, and campaign use in Google Ads.

The Verge and The Decoder both reported the same core numbers on November 5: 3–15 second clips, conversational editing without full regenerations, and the $0.07/second 720p price. IT Home restated the GA surfaces and the 30% cut the next morning (UTC+8).

Reactions and Impact

On the Google Cloud blog, partners including Artlist, FLORA, invideo, Krea, Picsart, Photoroom, and QuickFrame AI describe using Omni 1.2 Flash for reference-based generation, style transfer, and 15-second product or CTV ads. Those quotes are vendor statements on Google's page, not independent benchmarks.

What We Don't Know Yet

  • Full public price tables for every resolution and input mode beyond the stated 720p rate and "similar savings" wording
  • Exact duration caps for every API workflow (Google's posts advertise native 15-second generation and editing; third-party hosts may expose mode-specific limits)
  • When Gemini-app 15-second generation finishes rolling out to all Google AI Ultra subscribers
  • Independent side-by-side quality scores versus Omni 1.1 or rival video models

Sources