Hermes Vector

Tech

Google Releases Gemini Omni 1.2 Flash With 15-Second Video Generation

Google has released Gemini Omni 1.2 Flash, an AI video model that generates clips up to 15 seconds long at $0.07 per second for 720p, about 30% cheaper than its predecessor.

By Hermes-Vector AI Desk

Checked against 4 sourcesbasic-tutorials.compulse2.commediasat.infoforklog.com
Illustration for: Google Releases Gemini Omni 1.2 Flash With 15-Second Video Generation

In brief

  • Google released Gemini Omni 1.2 Flash on November 5, 2026, with clips up to 15 seconds long.
  • 720p video costs $0.07 per second, about 30% less than Omni 1.1.
  • The model is generally available in the Gemini API and Google Cloud.
  • In the Gemini app, access is initially limited to Google AI Ultra subscribers.

Google Releases Gemini Omni 1.2 Flash With 15-Second Video Generation

Google released Gemini Omni 1.2 Flash on November 5, 2026. The AI video model natively generates and edits clips up to 15 seconds long, up from 10 seconds previously, and a 720p video costs $0.07 per second via the API. That is about 30% less than with its predecessor, Omni 1.1. The model is now generally available in the Gemini API and on Google Cloud, and is gradually rolling out in Google Vids, Flow, the Gemini app, and Google Ads.

Key Features

Clips can run up to 15 seconds, with length adjustable in one-second increments starting at 3 seconds. A full 15-second 720p clip costs about $1.05. The model supports resolutions of 360p, 720p, 1080p, and 4K at 24 frames per second. Existing videos up to 15 seconds long can also be used as input for editing.

Users can control generation using images of characters, products, or style templates, along with short motion clips of up to three seconds and audio cues. The model is designed to maintain faces, clothing, text on packaging, and brand colors from the first frame to the last, even during camera cuts. Google demonstrates this with a virtual fitting example in which a 15-second 360-degree sequence is created from a model photo and front and back views of three outfits.

Multi-step editing in chat allows users to change the background, lighting, weather, product, or aspect ratio without regenerating the clip. This works over the full 15-second duration. Dialogue is intended to sound more natural, with cleaner lip-sync and more reliable voice assignment for multiple speakers. The model also promises long, continuous shots such as FPV drone flights, orbits, or dramatic zoom-outs with significantly fewer unwanted cuts.

Pricing and Availability

The API model ID is gemini-omni-1.2-flash. The previous model, gemini-omni-1.1-flash, will be redirected to the new model starting December 3, 2026.

Pricing per second:

  • 360p: approximately $0.023
  • 720p: $0.07
  • 1080p: approximately $0.104
  • 4K: approximately $0.209

A 15-second clip at 720p costs about $1.05. There is no free API access for video generation; in AI Studio, an account with active billing is needed.

Gemini Omni 1.2 Flash is generally available in the Gemini API and Google Cloud. For businesses, this marks the first-ever general availability release of Gemini Omni. Google Vids offers 15-second clips at no extra charge with a Google account or Workspace subscription, within Vids’ monthly AI limits. In the Gemini app, 15-second videos are initially rolling out only to Google AI Ultra subscribers. The model is also available in Flow and Google Ads.

All outputs carry an invisible SynthID watermark that verifies their AI origin.

Limitations and Industry Context

Google acknowledges that the model still struggles with fast-moving sports footage, rapid changes in scripted scenes, and complex physical interactions, and says work on those areas is ongoing. Interpolation between start and end frames, as well as extending an existing clip, still takes 3 to 10 seconds per step. The 15-second figure applies to normal generation, not to every workflow.

Businesses and creative organizations already experimenting with the technology include Adobe, Monks, Havaianas, and Whering. Adobe plans to incorporate the updated model into Firefly. Monks has been testing the model for visual effects and controllable video generation. Whering has evaluated the model’s capabilities for accurately representing clothing and fabrics in virtual styling applications.

Why it matters

The release extends native AI video generation from 10 seconds to 15 seconds while cutting the price by about 30%, making longer clips more affordable for advertisers, developers, and content creators. In-dialog editing lets users refine clips round by round instead of regenerating entire videos, which can save time and money. General availability in the Gemini API and Google Cloud gives businesses predictable capacity through Provisioned Throughput, while integration into Google Vids, Flow, the Gemini app, and Google Ads puts the model inside Google’s existing product ecosystem.


Sources

This article was drafted with AI assistance and checked against the sources above. Company claims are reported as claims. Cover image is AI-generated.

The day's checked news, by email

One short email a day with the top stories. Free, and you can unsubscribe in one click.

Or follow for the latest news:

Advertisement

How do you feel about this story?

Comments 0

No comments yet. Start the discussion.

Tech

Over 800 Artex AI Hacking Tools Found Exposed Online

Security firm Censys found more than 800 Artex deployments accessible without a password. The tool has been linked to attacks on South Korean finance firms and ConnectWise.

Checked against 6 sourcesTechCrunchtechgig.comthe420.in+3 more