01
In brief
Google released Gemini Omni Flash 1.1 (gemini-omni-1.1-flash) to general availability on August 27, 2026. It is the stable version of Google's fast conversational video generation and editing model and replaces the earlier gemini-omni-flash-preview, which Google says will be deprecated September 30, 2026.
02
What changed in Gemini Omni Flash 1.1
The GA release adds three headline controls: video extension, first-and-last-frame interpolation, and explicit output resolution. The model supports text, image and video input and outputs 3–10 second video at 24 FPS. Google lists a 1,048,576-token context window. Resolution options are 360p, 720p, 1080p and 4K; Google explicitly says 1080p and 4K are upscaled outputs rather than native high-resolution generation.
03
Video generation and editing
Gemini Omni Flash can generate video from text prompts, animate still images and edit video through the Interactions API. Its stateful editing workflow lets a developer generate a clip, then send follow-up natural-language instructions while retaining the previous interaction state. Google says the model processes text, image, audio and video as multimodal context.
04
Video extension
The new extend task can continue an existing clip from its ending. This is materially different from simply regenerating a longer prompt because the existing video becomes context for the continuation. Google's model card allows video input up to 10 seconds for editing and extension.
05
First and last frame interpolation
Developers can provide two images and ask Omni Flash to create the transition between them. This makes the model relevant for controlled camera moves, transformations, storyboard transitions and creator workflows where both endpoints matter.
06
Resolution control and 4K
The API's response format accepts 360p, 720p, 1080p and 4K, with 720p as the default. Treat 4K carefully in comparisons: Google's documentation identifies both 1080p and 4K as upscaled outputs.
07
How to access the API
The stable model ID is gemini-omni-1.1-flash. Google documents Python, JavaScript and REST examples through the Interactions API. Portrait 9:16 and landscape 16:9 aspect ratios are supported, which makes the model directly relevant to Reels, Shorts and other creator video workflows.
08
Pricing
Google's current paid-tier pricing lists $1.50 per million input tokens for text, image, video or audio, $9 per million text-output tokens, and $17.50 per million video-output tokens. Google calculates 720p video at 5,792 output tokens per second, which it says is an effective price of approximately $0.10 per second. There is no free API tier for this model in the current table. Higher-resolution cost follows actual output-token consumption; do not assume every resolution costs exactly the 720p effective rate.
09
Who should care
Creators gain a faster route from draft generation to conversational edits, portrait output, interpolation and higher-resolution delivery. Developers gain a stable model ID and API workflow instead of building around the preview endpoint. Marketing teams can use the combination of 360p drafts and higher-resolution final outputs as a prototype-to-delivery workflow, subject to current pricing and product policies.
10
Key limitations
Outputs are 3–10 seconds per generation according to the official model card. 1080p and 4K are upscaled. Availability, rate limits and pricing should be checked against current Google documentation before production use.
11
Bottom line
Gemini Omni Flash 1.1 matters less because it adds another AI video model and more because Google has moved a conversational video generation/editing workflow to a stable GA endpoint. The new extension, interpolation and resolution controls make it substantially easier to build practical creator and developer workflows around it.
Sources
Primary and supporting sources
Facts were rechecked against the linked sources immediately before publication. Pricing, product availability and rollout status can change.