Superpower Daily
ChatGPT Lets Temporary Chats Use Memory, Then Convert Them to R...Model-Serving Platforms Reach 6.1% of AI-Spending Businesses, W...Gemini Omni 1.1 Flash Adds 40-Second Scene Extensions and 4K Fi...Databricks Says 300M Chart-JSON Pipeline Tops Four Multimodal B...London Surgeons Remove 11mm Brain Tumour With Live AI Anatomy M...Nvidia’s $63.1 Billion Receivables Add a Cash-Flow Caveat to It...ChatGPT Lets Temporary Chats Use Memory, Then Convert Them to R...Model-Serving Platforms Reach 6.1% of AI-Spending Businesses, W...Gemini Omni 1.1 Flash Adds 40-Second Scene Extensions and 4K Fi...Databricks Says 300M Chart-JSON Pipeline Tops Four Multimodal B...London Surgeons Remove 11mm Brain Tumour With Live AI Anatomy M...Nvidia’s $63.1 Billion Receivables Add a Cash-Flow Caveat to It...
Modelspublished

Gemini Omni 1.1 Flash Adds 40-Second Scene Extensions and 4K Finishing Controls

Google is combining continuity controls, lower-cost previews and high-resolution finishing in one video workflow, though its extension and reference windows remain short.

By 3 min read
Gemini Omni 1.1 Flash Adds 40-Second Scene Extensions and 4K Finishing Controls
Sign in to save and react

Listen to this story

The audio brief

About 1:28
0:001:28
Read transcript
Google’s Gemini Omni 1.1 Flash is adding a more structured way to build video: generate a cheap draft, extend the scene in controlled steps, then finish at high resolution. The biggest change is scene continuation. The model can analyze up to ten seconds of earlier footage before creating the next segment, rather than relying only on the final second. Each extension adds ten seconds, and the full scene can reach forty seconds cumulatively. Google says that extra context is intended to improve visual consistency and adherence to the narrative. Developers can also define a shot’s first and last frame, letting the model generate the movement between them for transitions, camera orbits, zooms, or loops. A reference video can add visual context and character guidance, but that input is limited to three seconds. The workflow is designed to separate exploration from delivery. Google says 360p previews run up to sixty percent faster than standard 720p output and cost one-third as much. Listed pricing rises from three cents per second at 360p, to ten cents at 720p, fifteen cents at 1080p, and thirty cents at 4K. Higher-resolution upscaling is available for final output. The model is offered through Google AI Studio and the Gemini Enterprise Agent Platform, with access also rolling out in Google Flow and the Gemini app for AI Plus, Pro, and Ultra subscribers. The key constraint is still short memory: each continuation sees only ten seconds, and the complete extension stops at forty.

Story brief

3 key points

Google’s Gemini Omni 1.1 Flash turns video generation into a more controllable draft-to-finish workflow, with low-cost 360p exploration and paid upscaling to 4K. Its main constraint is also its differentiator: continuations use up to 10 seconds of prior footage per step and can extend a scene to 40 seconds total. Developers get keyframe interpolation and short reference-video inputs, while pricing rises from $0.03...

  1. 01

    Scene extensions add 10-second segments, capped at 40 seconds cumulatively; each continuation can analyze up to 10 seconds of earlier footage.

  2. 02

    First-and-last-frame controls support transitions, camera moves and loops; reference inputs can include up to three seconds of video.

  3. 03

    Google reports 360p previews run up to 60% faster and cost one-third as much as standard 720p, based on system throughput.

Google has released Gemini Omni 1.1 Flash as a production-ready generative-video update for developers. It adds scene continuation, first-and-last-frame transitions, video references, faster low-resolution drafts and upscaling, giving applications a more structured path from an early concept to final output.

A longer memory for continuing a shot

The central change is scene extension. Omni 1.1 can analyze up to 10 seconds of earlier footage before generating a continuation, instead of relying only on the final second. Developers can add footage in 10-second steps until a video reaches a cumulative 40 seconds. Google says the added context improves visual consistency and narrative adherence.

Keyframes and references define the next shot

Developers can specify a shot’s first and last frame, then have the model generate continuous video between those keyframes. Google positions that control for camera orbits, zoom transitions and looping clips. They can also supply up to three seconds of video reference material in multimodal input for visual context and character consistency.

Controls for draft and delivery

  • Generate 360p previews that Google says run up to 60% faster and cost one-third as much as standard 720p output.
  • Upscale generated video to 1080p or 4K for final output.

Resolution now carries an explicit price ladder

The listed per-second prices are $0.03 for 360p, $0.10 for 720p, $0.15 for 1080p and $0.30 for 4K. That creates a clear draft-to-finish tradeoff: explore a concept at lower resolution, then spend more only when moving to a higher-resolution output. Google bases its 360p speed comparison on system throughput.

Google puts the controls across its developer stack

Omni 1.1 is available through Google AI Studio and the Gemini Enterprise Agent Platform. It is rolling out globally in Google Flow for AI Plus, Pro and Ultra subscribers; scene extension is available in the Gemini app for those same tiers. Google also identifies Adobe, Figma Weave, GMI Cloud and Runway as users or integrations for Gemini Omni Flash in creative workflows.

Editorial analysis

Our Read

Our read: The important shift is the ordering of the tools. Google is pairing cheap 360p experimentation with keyframe transitions, short reference clips and later 4K upscaling, which points toward iterative creative workflows rather than a single finished generation. The 40-second extension ceiling and three-second reference limit make this a system for directing bounded shots and assembling outcomes. The next signal is whether platforms such as Adobe, Figma Weave and Runway make these controls durable editing primitives inside their products, rather than leaving them as isolated model settings.

Sources

  1. blog.googleGemini Omni 1.1 Flash lets you build with more control
  2. the-decoder.comGoogle's Gemini Omni 1.1 Flash makes AI video generation cheaper and more flexible

Related coverage

Continue reading