Google launches Gemini Omni 1.2 Flash with 15-second video generation and 30% lower pricing

Google launched Gemini Omni 1.2 Flash on 5 November 2026, making the multimodal video generation and editing model generally available across Google AI Studio, the Gemini API (gemini-omni-1.2-flash), Google Cloud Vertex AI (with Provisioned Throughput), Flow, Google Vids, Google Ads and the Gemini app for Google AI Ultra subscribers. Compared with Gemini Omni 1.1 Flash, Google cut standard 720p generation pricing by 30% to $0.07 per second.

The update extends native single-pass clip duration from 10 seconds in Omni 1.1 Flash to 15 seconds, configurable in 1-second increments from 3 to 15 seconds. Developers can draft clips at 360p or 720p, refine individual shots across multi-turn conversational prompts without regenerating from scratch, and upscale finished sequences to 1080p or 4K resolution. Every generated video carries an imperceptible SynthID watermark.

Reference conditioning and multi-turn editing

Gemini Omni 1.2 Flash accepts character, product and style reference images alongside short motion clips up to 3 seconds and 1-second voice samples, preserving faces, garment cuts, packaging typography and brand colours across 360-degree camera turns, first-to-last keyframe interpolations and continuous first-person-view (FPV) drone flythroughs. Multi-turn conversational edits let teams swap backgrounds, products, lighting, weather or aspect ratios across full 15-second clips while keeping character performances and camera framing intact, alongside multi-speaker dialogue routing and synchronised lip movement.

Key numbers

• $0.07 per second for 720p video generation, a 30% price reduction from Gemini Omni 1.1 Flash
• 3 to 15 seconds native clip duration in 1-second increments (up from 10 seconds in Omni 1.1)
• Up to 3-second motion reference clips and 1-second voice cues for multi-speaker lip-sync
• Draft generation at 360p or 720p with native post-generation upscaling to 1080p and 4K