Google DeepMind Debuts Gemini Omni 1.1 Flash for Pro Video
The production-ready update introduces granular creative controls, 4K upscaling, and a high-speed drafting mode to reduce generative video unpredictability.
Google DeepMind has released Gemini Omni 1.1 Flash, a production-ready update to its Omni model designed to bring professional-grade generative video capabilities to developers and creators. The update marks a strategic shift from experimental research toward a toolset optimized for integrated media editing software and creative workflows.
The new model introduces several advanced creative controls aimed at increasing precision. According to the Google Blog, scene extension now analyzes up to 10 seconds of prior context—a significant increase from the previous one-second limit—enabling users to extend videos in 10-second increments for a total duration of up to 40 seconds. To ensure smoother transitions and the creation of seamless loops, the model now supports the specification of both first and last frames to generate continuous video between two keyframes. Additionally, multimodal input now allows for the referencing of up to three seconds of video to maintain character consistency and visual context.
Production Efficiency and Scaling
To address the high cost and time requirements of video generation, Google has introduced a 360p draft mode. This mode is up to 60% faster and costs one-third as much as the standard 720p resolution, allowing for rapid prototyping before final rendering. For high-end production needs, the model supports upscaling outputs to 1080p and 4K resolution.
Industry Implications
These updates directly target the "uncontrollability" that has historically plagued generative video. By providing granular control over duration, keyframe transitions, and resolution, Google is positioning Gemini Omni 1.1 Flash as a viable component of professional production pipelines rather than a novelty tool. The introduction of the low-cost drafting mode specifically lowers the barrier for iterative design, allowing creators to refine concepts quickly without exhausting computational budgets.
Availability and Deployment
Gemini Omni 1.1 Flash is now available through several Google channels. Developers can access it via the Gemini API in Google AI Studio and the Gemini Enterprise Agent Platform. For end-users, the capabilities are available to Google AI subscribers within the Gemini app and Google Flow. Future developments will likely focus on further expanding the maximum duration of generated clips and refining the consistency of long-form multimodal inputs.