Announced 5 Nov 2026 · Sources checked
What Gemini Omni 1.2 Flash brings
Google announced Gemini Omni 1.2 Flash on November 5, rolling it out as a generally available model across the Gemini API, Google Cloud's Vertex AI, and consumer products including Flow, Google Vids and the Gemini app. The headline upgrade is native 15-second video generation, up from 10 seconds in Omni 1.1, with support for specifying duration in one-second increments from 3 to 15 seconds.
According to Google DeepMind's announcement, the model introduces multi-turn conversational video editing, meaning users can refine the same clip turn by turn—swapping a product or background, changing lighting or weather conditions, or switching the aspect ratio—without re-generating the entire clip. This workflow is designed to let creators prototype at lower resolutions then upscale their best takes to 1080p or 4K.
Price drops 30 percent from Omni 1.1 to $0.07 per second for 720p output. Google says the workflow keeps costs manageable by letting users draft fast at 360p or 720p, fix details with follow-up edits instead of re-generating, and upscale only the final selects.
Better reference adherence and consistency
A major focus of the 1.2 release is consistency. The model can take character, product or style images, short motion clips up to three seconds, and audio cues as references and hold onto those details across the full generation. Google says faces, garments, packaging typography and brand colors stay consistent across camera angles in generated clips.
Demonstrations shared by Google include a virtual try-on sequence where a single model photo and front-and-back product shots of three outfits produce a 15-second 360-degree turn while keeping the model's identity and each garment's fit and back details consistent across match cuts. Another demo shows characters' performances, armor reflections and camera framing staying stable across five background swaps in a multi-turn editing session.
Multimodal input and cinematic output
Gemini Omni 1.2 Flash takes combinations of images, audio, video and text as input and generates videos grounded in Gemini's real-world knowledge. Google says the model now handles more complex, continuous camera movements—sweeping first-person flythroughs, dramatic zoom-outs and orbits—with fewer jump cuts and steadier framing.
The model also improves natural speech in generated scenes, with better lip-sync and distinct multi-speaker conditioning. This applies across Flow, Google Vids and Gemini for business, where users can create multi-character dialogue scenes, video explainers and polished videos with avatars, music and narration. Outputs carry an imperceptible SynthID watermark.
Availability and rollout
Gemini Omni 1.2 Flash is rolling out immediately across Google's product lineup. Developers can access it through the Gemini API and Google AI Studio, with starter apps and documentation available. Businesses can use it via the Google Cloud console with enterprise-grade infrastructure and Provisioned Throughput.
For consumers, Google is enabling 15-second clip generation at no additional cost in Google Vids with any Google account or Google Workspace subscription. The Gemini app for AI Ultra subscribers will also gain 15-second video creation capabilities as the rollout progresses. Google says development continues on areas including fast sports action, rapid script changes and complex physics simulation.
How the model compares to its predecessor
The upgrade from Omni 1.1 to 1.2 Flash represents a 50 percent increase in maximum generation length, from 10 to 15 seconds, alongside the 30 percent price reduction. Google demonstrated side-by-side comparisons showing Omni 1.2 Flash adhering more closely to reference images than 1.1 Flash, particularly in multi-layer scene composition tasks where the model maintained consistent materials, shadows and typography through an entire 15-second clip.
Common questions
How much does Gemini Omni 1.2 Flash cost?
The model costs $0.07 per second for 720p video output, a 30 percent reduction from Omni 1.1. Google’s workflow lets users draft at 360p or 720p, edit, and upscale only final selects to 1080p or 4K to keep costs down.
Can I generate longer videos by chaining clips?
The model’s native limit is 15 seconds per generation. Google says you can extend a clip by interpolating between a first and last keyframe, building longer sequences while keeping products, characters, props and environments consistent from shot to shot.
Is Gemini Omni 1.2 Flash available for free?
Free-tier users can generate 15-second clips in Google Vids with a Google account or Google Workspace subscription. The full API and business access is priced per-second. The Gemini app for AI Ultra subscribers will also get access as the rollout progresses.
Sources & further reading
How this story was made
Written by Kristian Kostov with AI assistance and checked against the linked sources. Company performance claims are attributed to the company. Analysis reflects AiLookout’s interpretation; we have not independently tested the products discussed. The cover is an AI-generated editorial illustration, not a photograph or evidence of an actual event.
Our editorial standards
