Gemini Omni 1.1 Flash lets you build with more control
If you create marketing, training or social content, this makes AI-generated video faster and cheaper to draft (start rough at 360p, only pay full price once you're happy) and gives you more creative control over how clips flow — a practical upgrade if you're already using Gemini or Flow for video.
What changed
Source summary
Google released Gemini Omni 1.1 Flash, an upgraded video-generation model with scene extension (videos up to 40 seconds), first/last-frame keyframe control for camera movement, faster/cheaper 360p drafting, and up to 4K upscaling, available via Google AI Studio, the Gemini API, Flow, and the Gemini app.
Why it matters to your work
Practical interpretation
If you create marketing, training or social content, this makes AI-generated video faster and cheaper to draft (start rough at 360p, only pay full price once you're happy) and gives you more creative control over how clips flow — a practical upgrade if you're already using Gemini or Flow for video.
When this could help
Worth trying now if you are building video-generation features or need more control than single-shot generation offers — especially scene extension for longer sequences and 360p drafts for faster iteration.
Gemini Omni 1.1 Flash adds scene extension with up to 10 seconds of prior context (up from 1 second), first-and-last-frame interpolation for camera orbits and looping clips, 360p draft previews that render about 60% faster at roughly a third of 720p cost, 4K upscaling to 1080p or 4K, and up to 3 seconds of video reference input for visual consistency.
How to get started
Use model ID gemini-omni-1.1-flash in the Gemini API. For scene extension, pass previous_interaction_id. Set resolution via response_format (e.g. '360p' for draft previews).
Provider guidance ↗
Build it into your routine
SCALA suggested workflow
Iterate with 360p drafts
Generate 360p draft previews first to iterate on framing and motion at lower cost, then upscale to 720p or 4K once the composition is right.
What to check
- Scene extension analyses up to 10 seconds of prior context. Videos can be extended in 10-second increments to a cumulative maximum of 40 seconds. Video references accept up to 3 seconds. Source ↗
- Pricing is shown in a table image in the blog post but not as text — exact figures are not extractable from the announcement. Source ↗
Sources & provider guides
Official documentation checked 12 Oct 2026. SCALA prompts and workflows are suggestions, not claims of hands-on testing.
Source & availability
This is a summary of a published provider announcement, with practical context from SCALA’s feed.
Plan, region, workspace and rollout conditions may affect access in Australia. Check the original source for current availability.
blog.google