# Google Gemini Omni 1.1 Flash Adds Keyframe Controls and Extended Scene Context > The update equips developers with tools for incremental video building and resolution flexibility, advancing production workflows in a competitive AI media landscape. *Published 2026-08-28 · By Marcus Vance* Gemini Omni 1.1 Flash is a high-performance multimodal model for fast, conversational video generation and editing via the Gemini API. The release of Gemini Omni 1.1 Flash represents Google's response to demands for greater precision in AI video tools. Developers have faced difficulties with short context windows that produced disjointed extensions and limited creative direction. This model incorporates longer context retention and frame specification to support more coherent outputs. Integration with natural language allows iterative refinements without specialized interfaces. Such changes aim to move AI video from experimental to practical use in professional settings. ## What background led to the development of Gemini Omni 1.1 Flash? AI video generation has progressed quickly yet often struggled with maintaining visual consistency across extended sequences. Earlier models restricted reference to only the final second of existing footage, which frequently resulted in abrupt shifts in motion or lighting. Google DeepMind addressed this by expanding the reference window to ten seconds of prior context. This adjustment enables incremental building of content in ten second segments that accumulate to forty seconds total. The approach draws on Gemini's established multimodal reasoning to handle text, image, and video inputs within a single conversation thread. Competition in the space, including offerings like Veo, has pushed providers toward features that support end to end production pipelines. Creative teams require options for prototyping at low cost before committing to high resolution renders. The addition of draft modes and upscaling pathways directly targets these workflow needs. Access through the Gemini API further allows embedding into existing development environments such as Google AI Studio. ## What new features define Gemini Omni 1.1 Flash in detail? Scene extension now draws on up to ten seconds of preceding video rather than one second, producing more fluid continuations. Users define starting and ending frames to direct camera paths and scene transitions. This control reduces the need for post generation corrections. The model accepts text prompts alongside images and short video clips for editing tasks. Natural language conversations via the Interactions API permit ongoing refinements to generated clips. Resolution flexibility spans draft quality at 360p through standard 720p and 1080p up to 4K final output. All options maintain 24 frames per second. The lower resolution mode supports rapid iteration at lower expense. Once a draft meets approval, the system can upscale it while preserving motion details. These options collectively reduce barriers for teams that alternate between quick tests and polished deliveries. ## How do the technical specifications support varied use cases? Input handling covers text descriptions, static images, and video segments limited to ten seconds for editing or extension operations. The context window reaches 1,048,576 tokens to accommodate complex multimodal prompts. Output clips range from three to ten seconds per generation request. Model code gemini-omni-1.1-flash identifies the specific variant within the Gemini API documentation. These parameters enable both short experimental clips and building toward longer sequences through repeated extensions. Supported output resolutions and associated use cases for Gemini Omni 1.1 FlashResolutionFrame RatePrimary UseRelative Cost360p24 FPSDrafting and prototypingLowest720p24 FPSStandard generationMedium1080p24 FPSHigh quality editingHigher4K24 FPSFinal production outputHighest ## What workflow steps enable extended video creation? - Initiate generation with text or image prompts via the Gemini API. - Provide up to ten seconds of prior video context for extension analysis. - Define first and last frames to control camera movement and transitions. - Generate initial output in 360p draft mode for rapid review. - Upscale approved content to 1080p or 4K for final delivery. This sequence allows users to prototype quickly before investing in higher fidelity renders. The conversational interface supports adjustments at any stage through plain language requests. Integration with platforms such as Figma Weave further streamlines version management and reference attachment across team members. ## What market and stakeholder implications follow the release? The model positions Google to compete more effectively in enterprise and creative markets where controllable video tools hold value. Reduced iteration costs through draft modes may accelerate adoption among smaller studios and independent developers. Larger production houses gain pathways to incorporate AI assistance without replacing established pipelines. API availability encourages third party applications that embed these capabilities into specialized software. Stakeholders in design and marketing sectors benefit from the shift toward directed generation rather than one shot outputs. Consistent editing through language inputs lowers training requirements for teams. Over time, such features could standardize AI involvement in video production cycles across industries. ## How have experts reacted to Gemini Omni 1.1 Flash? Industry feedback emphasizes the practical advantages for collaborative creative work. Partners note improved ability to iterate on generations while maintaining visual coherence. The combination of extension tools and high resolution options supports a move from basic generation to full direction of content. > Gemini Omni Flash is one of the strongest video models available in Figma Weave, where the canvas helps creative teams build on every generation — attaching references, branching different versions, and shaping something unique. With extensions, richer reference material, and 4K resolution, Gemini Omni Flash takes teams beyond generating videos to truly directing them.Itay Schiff, Creative Director, Figma Weave These observations align with broader trends toward AI systems that function as collaborative partners rather than autonomous generators. Continued refinement of context handling is expected to further enhance reliability in longer form projects. ## What developments can be anticipated next? Future iterations may extend context windows beyond ten seconds or introduce additional frame level controls. Enhanced integration with other Google services could streamline end to end production from script to final render. Competition will likely drive further gains in speed, consistency, and cost efficiency for multimodal media models. Developers should monitor API updates for new parameters that expand creative flexibility. The emphasis on conversational editing suggests ongoing investment in reasoning layers that interpret complex user intent across video tasks. ## Sources 1. [Details on scene extension with 10 seconds of context, frame specification, 4K outputs, and the 60% faster one third cost statistic for 360p mode.](https://blog.google/innovation-and-ai/technology/developers-tools/build-with-gemini-omni-1-1-flash/) 2. [Model specifications including gemini-omni-1.1-flash code, input types, output resolutions from 360p to 4K at 24 FPS, context window size, and support for video extension and upscaling.](https://ai.google.dev/gemini-api/docs/models/gemini-omni-flash) --- Source: https://aiintelreport.com/frontier-models/google-gemini-omni-1-1-flash-video-release Index: https://aiintelreport.com/llms.txt · Full text: https://aiintelreport.com/llms-full.txt