Google Releases Gemini Omni 1.2 Flash Video Model with Native 15-Second Generation and 30% Price Cut

Original title: AI 视频模型卷出新高度:谷歌 Gemini Omni 1.2 Flash 发布,原生 15 秒生成、较 1.1 降价 30%
78AI Score

Google has launched Gemini Omni 1.2 Flash, a video generation model now generally available on Gemini API and Gemini for business. It extends native clip length from 10 to 15 seconds, adds multi-turn conversational editing and stronger reference consistency, and cuts 720p pricing to $0.07 per second, 30% below version 1.1.

Full Report

IT之家 November 6 news: Google has officially launched the Gemini Omni 1.2 Flash video generation model, making it fully available on the Gemini API and Gemini for business platforms.

The new model achieves native 15-second video generation, multi-turn conversational editing, and stronger reference adherence, while also being 30% cheaper than version 1.1. The price for 720p video generation is $0.07 per second (IT之家 note: approximately 0.47 RMB at current exchange rates).

Google Gemini Omni 1.2 Flash released

In terms of cost reduction at scale, the new workflow allows users to first quickly draft at 360p or 720p, then correct details through subsequent edits without regenerating the entire clip, and finally upscale the finished product to 1080p or 4K, further reducing usage costs.

In terms of clip length, Gemini Omni 1.2 Flash extends the natively supported video duration from 10 seconds in version 1.1 to a maximum of 15 seconds, and supports custom durations in 1-second increments within the 3 to 15 second range, enabling the one-shot generation of complete stories, product explainers, or social advertising clips.

In terms of reference adherence, users can guide generation with character, product, or style images, motion clips of up to 3 seconds, and audio cues. The model can maintain consistency in faces, clothing, packaging typography, and brand colors across different camera angles throughout the entire process, making it easy to turn existing product images and character design sheets into virtual try-ons, product turntable displays, and branded video ads.

In terms of output consistency, the new model can maintain strong consistency of characters, objects, and scenes from the first frame to the last frame, reducing visual drift and distortion during motion and camera cuts. Whether interpolating between first and last keyframes or extending clips to build long sequences, products, characters, props, and environments remain consistent across different shots.

In terms of conversational editing, users can adjust the same clip turn by turn—for example, replacing products or backgrounds, modifying lighting or weather, or switching aspect ratios—without regenerating from scratch. This multi-turn conversational editing capability covers the full 15-second clip, making it convenient to generate ad variants for different audiences and placements without rewriting prompts or regenerating.

In terms of speech performance, the new model can generate smoother and more natural dialogue and clearer lip synchronization, and supports multi-speaker conditional control. It is suitable for directing multi-character dialogue in Flow, producing explainer videos in Gemini for business, and creating professional videos with digital humans, music, and voiceover in Google Vids.

In terms of cinematic visual control, users can direct complex continuous camera movements, such as panoramic FPV fly-throughs, dramatic pull-backs, and orbiting shots, greatly reducing unnecessary jump cuts and maintaining more stable scaling and a more refined cinematic quality.

Gemini Omni is now fully generally available for the first time. Backed by Google Cloud's enterprise-grade infrastructure and security capabilities, it supports Provisioned Throughput, and generated content carries an imperceptible SynthID watermark.

Developers can prototype and debug, view documentation, and check pricing through Google AI Studio; enterprise users can try it in the Google Cloud console; general users can direct scenes and perform multi-turn editing in Flow, generate 15-second clips for free in Google Vids, and access the Gemini App, which is gradually rolling out to Google AI Ultra subscribers, while Google Ads supports scaled video ad delivery.