Gemini Omni 1.1 Flash in Genosai — Google video from 360p drafts to 4K

Gemini Omni 1.1 Flash is Google DeepMind's production video model, available in Genosai. It generates 4, 6, 8 and 10-second clips from text, accepts up to 7 reference images plus a video input for character consistency, and its resolution range covers the whole workflow: fast 360p drafts, 720p standard, and 1080p or 4K finals. In Genosai the model runs online — no Google developer account, no API setup.

Updated: August 31, 2026

Contents

What is Gemini Omni 1.1 Flash

Gemini Omni 1.1 Flash is a video model by Google DeepMind, presented on August 27, 2026 as the production release of the Gemini Omni line. The Flash suffix is Google's label for fast workhorse models: the focus is not record-setting demos but predictable results inside a pipeline. The launch partner list says it all: Adobe built the model into Firefly for video editing, Figma into Weave, and Runway announced support as well. Figma's wording captures the release's idea precisely — creative teams want to "direct videos, not just generate them" — hence the emphasis on references and control.

The model covers the full production cycle of resolutions. According to Google, 360p drafts render about 60 percent faster than standard 720p, so ideas and storyboards iterate quickly. The standard is 720p, and for final production there are 1080p and true 4K. Character and style consistency rests on multimodal references: up to 7 images plus a video input in a single generation.

In Genosai the model runs online: no Google developer account, no API keys, no foreign card required. Open the video studio in the browser, pick the model, describe the scene — and download the clip a couple of minutes later.

Capabilities

Here is what Gemini Omni 1.1 Flash can do in Genosai in practice.

360p drafts and 4K finals

Four resolutions in one model is rare among video generators. The workflow logic: composition, pacing and the prompt get tuned on fast 360p drafts — which in Genosai cost the same as 720p but finish noticeably sooner. The final version is rebuilt with the same prompt in 1080p or 4K: pricier, but only once. For footage headed to a big screen or an ad account, a true 4K render matters.

Up to 7 image references

References are the main control tool. Images define the character's look, the product, the palette and the composition of key frames: the model treats them as the scene's visual vocabulary. The slots are shared: there are 7 in total, and a regular image takes one. For brand content this is how you keep the product recognizable instead of gambling on a text description.

A video input for context

Beyond images, the model accepts one video clip up to 30 seconds — the first up to 10 seconds go into the generation. A video input conveys what is hard to put into words: the character's movement style, the camera's character, the location's atmosphere. It occupies two of the seven slots and is billed at a flat per-generation price — 168 credits, or 252 in 4K, regardless of the output duration.

Predictable durations

Four fixed options — 4, 6, 8 and 10 seconds — each with a fixed price. No per-second arithmetic: the cost of the clip is known before launch and does not depend on what happens in the frame. Aspect ratios are 16:9 for horizontal platforms and 9:16 for Reels, Shorts and vertical feeds.

The Omni ecosystem

Google's announcement also lists scene extension — up to 40 extra seconds with 10 seconds of context taken into account — and first-and-last-frame generation. These modes are not yet exposed in Genosai, but the model itself is the same one running under the hood of Adobe Firefly and Google Flow: the image quality is identical.

Examples and prompts

Below are real generations: the prompt and the clip Gemini Omni 1.1 Flash produced from it in Genosai on the first try. All three examples are fast drafts in 360p, the model's cheapest and quickest mode.

A cinematic wildlife scene, 16:9, 8 seconds:

A curious red fox trots through a snowy forest at golden hour, breath visible in the cold air, snowflakes sparkle as they fall, cinematic tracking shot, shallow depth of field, warm rim light.

A fox in a winter forest, 16:9, 8 seconds — generated in Genosai.io with Gemini Omni 1.1 Flash

A vertical macro clip for social feeds, 9:16, 6 seconds:

Vertical social clip: a barista pours latte art in slow motion, milk swirls into a rosetta pattern, steam rises, cozy cafe bokeh lights in background, top-down to side angle camera move.

Latte art, 9:16, 6 seconds — generated in Genosai.io with Gemini Omni 1.1 Flash

A stylized 3D animation, 16:9, 4 seconds — the minimum 63-credit tier:

A tiny cute robot waters a sunflower on a sunny windowsill, the flower turns toward the light, soft pastel colors, 3D cartoon style, gentle camera push-in.

A robot and a sunflower, 16:9, 4 seconds — generated in Genosai.io with Gemini Omni 1.1 Flash

Write prompts in English and structure them as subject, action, setting, camera, light. The model understands cinematography terms well: tracking shot, push-in, shallow depth of field. When using references, let the text prompt play director — what happens and how the camera moves — while the images define the look and style.

How to use it in Genosai

  1. Sign in to Genosai and open the video studio.
  2. Pick Gemini Omni 1.1 Flash from the model list.
  3. Describe the scene; upload up to 7 reference images or a video clip if needed.
  4. Choose the aspect ratio, resolution and duration — the price updates on the button instantly.
  5. Start the generation and download the finished clip.
  6. For the final, re-run the best draft in 1080p or 4K with the same prompt.

Pricing

The price is fixed per clip and depends on resolution and duration. At 360p, 720p and 1080p a four-second clip costs 63 credits, six seconds — 84, eight — 105, ten — 126. In 4K: 147, 168, 189 and 210 credits respectively. Generation with a video input has a flat price: 168 credits at any resolution up to 1080p and 252 in 4K. The exact amount is always shown on the button before launch, and current credit packages are on the Pricing page. Free starter credits let you try the model after signing up.

Comparison

Versus the older Gemini Omni Video: 1.1 Flash is the newer production release with the 360p draft mode and refreshed quality, and both cards work identically in the studio — prompt migration is seamless. Start with 1.1 Flash today.

Versus Veo 3.1 Fast, Google's other video model in Genosai: Veo generates audio and speech but caps at 8 seconds, from 60 credits per clip and with no draft resolution. Omni 1.1 Flash wins on control: references, a video input, four resolutions and a 10-second option — but no sound.

Versus Wan 3.0 and Seedance 2.5: both make 30-second scenes with audio, billed per second. For a long clip with a soundtrack they are the obvious choice. For a short, precisely-directed clip with a reference-locked character, fast draft iterations and a 4K final — that is the niche of Gemini Omni 1.1 Flash.

Limitations and tips

The model does not generate audio — add the soundtrack in editing, or pick another model when sound is critical. Of the capabilities Google announced, scene extension and the first-and-last-frame mode are not yet exposed in Genosai; references work in full. The input quota is shared: images and video split 7 slots, video takes two — the studio will not let you exceed the limit.

Practical tips: iterate at 360p — it is the model's fastest mode, and the price matches 720p; build the final with a separate 1080p or 4K run. References work best when each one owns a single job: the character separately, the product separately, the style separately. Choose the video input when a motion style is easier to show than to describe. And mind the moderation: real faces and copyrighted characters in references are blocked — use fictional heroes.

FAQ

What is Gemini Omni 1.1 Flash and who developed it?

Gemini Omni 1.1 Flash is Google DeepMind's production video model, announced on August 27, 2026 as the next step of the Gemini Omni line. It targets real production pipelines: multimodal references, fast 360p drafts and output up to 4K. In Genosai it is available online with no API setup.

How is 1.1 Flash different from Gemini Omni Video?

It is the newer, faster production release of the family. The key additions are a 360p draft mode that renders roughly 60 percent faster than standard and expanded reference capabilities. In Genosai both models live as separate cards with identical form mechanics.

Which durations and resolutions are supported?

Durations are 4, 6, 8 and 10 seconds. Resolutions are 360p for drafts, 720p as the standard, and 1080p or 4K for final renders. Aspect ratios are horizontal 16:9 and vertical 9:16.

How many references does the model accept?

Up to 7 images per generation. A video reference takes two of the seven slots, so alongside a clip you can pass up to 5 images. References are how you lock the character's look, the product and the scene style.

How does the video input work?

You upload a clip up to 30 seconds long, and the model takes its first up to 10 seconds as visual context — the character, the setting, the motion style. Generation with a video input has a flat price: 168 credits, or 252 in 4K.

How much does a generation cost in Genosai?

The price depends on resolution and duration: at 360p, 720p and 1080p — 63, 84, 105 and 126 credits for 4, 6, 8 and 10 seconds respectively; in 4K — from 147 to 210 credits. The exact amount is always shown on the button before you start.

Does the model generate audio?

No — Gemini Omni 1.1 Flash focuses on the visuals and does not generate a soundtrack. If you need a clip with sound and speech in one pass, look at Wan 3.0 or Seedance 2.5 in the same video studio.

Try Gemini Omni 1.1 Flash on Genosai