Clarity for AI buildersSponsored by AI Routing
pricing

Google rolls out Gemini Omni 1.2 Flash with 15-second video and lower pricing

Google rolls out Gemini Omni 1.2 Flash with 15-second video and lower pricing
Google DeepMind is making its Gemini Omni 1.2 Flash video-generation model generally available, with clips up to 15 seconds long, conversational editing and a 30 percent price reduction, according to The Decoder’s report on Google’s announcement. The model supports clips between three and 15 seconds, up from a previous maximum of 10 seconds. Pricing for 720p video falls to $0.07 per second. Google says the rollout spans the Gemini API, Google AI Studio, the Google Cloud Console, Flow, Google Vids, the Gemini app for Google AI Ultra subscribers, and Google Ads. Every generated video carries an imperceptible SynthID watermark indicating its origin. Conversational editing lets users adjust a generated clip through successive requests, changing elements such as the background, lighting or aspect ratio without regenerating the entire clip. According to Google, the feature works across the full 15-second duration. The company describes a tiered workflow intended to reduce production costs: create drafts at 360p or 720p, correct individual details through follow-up edits, then upscale only the final result to 1080p or 4K. The reported $0.07-per-second price applies specifically to 720p video; the supplied report does not provide pricing for other resolutions or editing operations. Google also claims improved adherence to reference material. Supported inputs include character images, product photos, motion clips of up to three seconds, and audio cues. The company says the model better preserves faces, clothing, packaging typography and brand colors across camera angles. In one example described by The Decoder, Google demonstrates a virtual try-on sequence generated from a single model photo and product shots of three outfits. The 15-second clip includes 360-degree turns while, according to Google, preserving the model’s identity and the garments’ cut and drape. Other claimed improvements include reduced visual drift and morphing during motion and camera transitions, more natural-sounding dialogue, clearer lip-sync and better distinction between multiple speakers. Google also says the model handles continuous camera movements, including FPV drone flights and dramatic zoom-outs, with fewer unwanted jump cuts. These quality improvements are Google’s claims as reported by The Decoder. The supplied material does not include independent testing or quantitative benchmarks to establish their extent.