Google Releases Gemini Omni 1.1 Flash with 40-Second Video Extensions and 4K Upscaling
On August 27, 2026, Google officially released Gemini Omni 1.1 Flash into General Availability across the Gemini API and Google AI Studio, expanding its multimodal video generation and editing capabilities for developers.
The updated model introduces a 10-second video context window that allows consecutive generation calls to chain into coherent clips up to 40 seconds long. For shot planning and storytelling, Gemini Omni 1.1 Flash adds first-and-last-frame conditioning, enabling creators to define precise start and end frames while the model generates intermediate temporal transitions. Additional controls include style reference matching and localized inpainting.
Google has structured API access around a tiered draft-and-upscale pricing model. Developers can generate initial 360p drafts at $0.03 per video second, preview iterations rapidly, and apply optional high-resolution passes up to 4K output. On public evaluations, Google reports that the model achieved 1,515 points on the crowdsourced Text-to-Video Arena leaderboard.
The release complements Google's text-oriented frontier lineup, including Gemini 3.7 Flash and the earlier Gemini 3.6 Flash, giving development teams programmatic video synthesis tools that integrate directly alongside conversational and multimodal agent workflows.
