Gemini Omni Flash 1.1

Gemini Omni Flash 1.1 is a high-performance multimodal video generation and editing model developed by Google DeepMind. Designed for real-time creative workflows, the model natively processes text, images, audio, and video simultaneously, enabling cohesive generation with synchronized native audio. It serves as an upgrade over earlier preview versions, offering granular production-ready controls through the Interactions API.

Core Capabilities and Controls

The model introduces several major architectural and feature enhancements for video production. Users can leverage scene extension to stretch video generations up to 40 seconds using a 10-second contextual window for improved narrative flow and consistency. Additionally, first and last frame interpolation allows creators to pin specific starting and ending points, ensuring smooth camera movements and exact structural transitions.

Performance and Output Modes

Gemini Omni Flash 1.1 supports flexible output pipelines, including a low-cost 360p draft mode for rapid prototyping and iterative prompt testing. Final projects can be exported with upscaled resolutions reaching 1080p and 4K, providing polished details for professional workflows. Through conversational editing, users can iteratively refine clips using natural language commands without needing to re-upload base files.