Google AI Releases Gemini Omni 1.1 Flash Video Model

Google AI has released Gemini Omni 1.1 Flash, a significant production update to its native multimodal video generation and editing model. This release transitions Omni from a basic generator to a more directly controllable tool for video creation. A key enhancement is the extended scene generation capability, which can now read up to 10 seconds of prior context, a substantial improvement over the previous single final frame reference. This allows for more coherent and extended video sequences. Furthermore, users gain precise control over camera movement by pinning the first and last frames of a scene, enabling specific directorial intentions to be realized within the generated video. The model also introduces significant efficiency improvements for draft rendering, producing content at 360p resolution at one-third the cost of 720p rendering, making iterative development faster and more economical. Final outputs can be upscaled to a high-resolution 4K, ensuring professional-grade quality. Video clips can now be used as references to maintain character consistency across different generated segments, a crucial feature for narrative storytelling and character-driven content. Gemini Omni Flash is built upon three core properties that distinguish it from earlier video generation models: native multimodality, allowing for the simultaneous processing of text, image, audio, and video inputs; conversational editing facilitated through the Interactions API, which enables iterative refinement of generated content; and the inheritance of extensive world knowledge from the broader Gemini family of models. The editing process is stateful, meaning users can pass a previous interaction ID to the model. The model then applies the new changes while preserving elements not explicitly mentioned in the prompt, eliminating the need to re-upload the entire prior video for minor adjustments. This stateful editing significantly streamlines the workflow for complex video projects. Gemini Omni 1.1 Flash is accessible via the Gemini API within Google AI Studio and the Gemini Enterprise Agent Platform. Several prominent industry players, including Adobe, Figma Weave, GMI Cloud, and Runway, have already been named as production users, indicating strong industry adoption and integration of the new model's capabilities. This release signifies a notable advancement in AI-powered video production, offering greater control, efficiency, and quality to creators.