Model · Google
Gemini Omni Flash 1.2, 1.1 and V1: Google AI video
Gemini Omni Flash is a family of Google video models: from text, a start image or reference images it produces a video shot, and it can also edit an existing video. Its newest version, 1.2, outputs 3 to 15 seconds from 360p to 4K, with sound; two companions follow it, Extend, which adds a continuation to a video, and Upscale, which raises its resolution. Versions 1.1 and V1 stay in the catalogue. On Beemm Vision they all run in Generate and on the canvas, on the same credit balance as the rest of the catalogue.
What Gemini Omni Flash brings
Four ways to start
Text alone; a start image, with a last frame if you want to pin the ending; up to ten reference images to hold a character or a set, plus three reference clips on the canvas; or a video of up to 15 seconds to edit with a prompt. In Generate these are the Frames, Reference, Text and Edit types of version 1.2.
3 to 15 seconds, 360p to 4K, with sound
Version 1.2 sets its length to the second between 3 and 15 (10 at most with a last frame), its resolution at 360p, 720p, 1080p or 4K, its format at 16:9 or 9:16. The soundtrack is generated with the picture, with no option to switch on, and the prompt can run to 20,000 characters.
Extend and upscale within the same family
Extend adds 3 to 10 seconds to an existing video, picking up from its last ten seconds; Upscale raises a video of up to 15 seconds to 720p, 1080p or 4K, with no prompt. Both accept a video made by another model, and the video upscaling guide compares Upscale with the catalogue's two other upscaling models.
Three versions, one picker
1.2, 1.1 and V1 are picked on the Version axis of the same line, mid-flow. 1.1 keeps the four modes over 3 to 10 seconds, 360p to 4K; V1, the original version, stays in the catalogue for anyone who wants its look, with text, start image and references.
The Gemini Omni Flash versions available here
Five entries, filed under Google in the Generate video picker (Version axis: 1.2, 1.1, 1.0) and offered as video nodes on the canvas.
- Gemini Omni Flash 1.2Text, start image (and last frame), references or a video to edit, to video. 3 to 15 s (10 s with a last frame), 360p to 4K, 16:9 or 9:16, sound included. Up to 10 reference images, plus 3 reference clips on the canvas; edits a video of up to 15 s.
- Gemini Omni Flash 1.2 · ExtendAdds 3 to 10 s (8 by default) to a video of up to 60 s, picking up from its last 10 seconds; prompt required, 360p to 4K.
- Gemini Omni Flash 1.2 · UpscaleRaises a video of up to 15 s to 720p, 1080p (the default) or 4K, above its original resolution; no prompt.
- Gemini Omni Flash 1.1Text, start image (and last frame), references or a video to edit, to video. 3 to 10 s, 360p to 4K, 16:9 or 9:16; up to 10 reference images, plus 3 reference clips on the canvas.
- Gemini Omni Flash V1Shown as “1.0” in Generate. Text, start image or up to 10 reference images to video, 3 to 10 s, 16:9 or 9:16; no last frame, no resolution setting, no video editing.
What Beemm Vision adds
A video is prepared upstream and reworked downstream — in the same graph, on the same balance.
- Prepare the start image with Nano Banana, Google's image family, then animate it with Gemini Omni Flash without an export.
- Compare Gemini Omni Flash, Kling, Seedance and MiniMax H3 on the same shot before choosing.
- Wire the keeper into Extend or Upscale, without downloading it.
- Launch a generation from Claude or ChatGPT with the MCP connector.
- Ship the graph as an app so the team can rerun it without opening the canvas.
- One credit balance for the whole catalogue: no Google Cloud account, no API key.
Gemini Omni Flash in practice: free use, extending, upscaling
Gemini Omni Flash for free: what the free plan covers
The Beemm Vision free plan opens the whole catalogue, Gemini Omni Flash included, with 100 free credits. It is not unlimited video: each generation spends credits according to the version, length and resolution you choose, and the exact amount is shown before you launch. What it does let you do is try the model on your own shots and compare it with the other models of the AI video generator before deciding anything.
Extending a video with Gemini Omni Flash
This is the Extend type of version 1.2. You attach the video to continue — 60 seconds at most — and describe what comes next: the model picks up from its last ten seconds and adds 3 to 10 seconds, 8 by default, at 360p, 720p, 1080p or 4K. The prompt is required, and only the added segment is billed. On the canvas, the output of a Gemini Omni Flash, Kling or Seedance node wires straight into the extension node.
Upscale: raising a video to 1080p or 4K
The Upscale type of version 1.2 raises a video of up to 15 seconds to 720p, 1080p — the default — or 4K, with no prompt. The target has to exceed the original resolution: 720p or 1080p from 360p, 1080p or 4K from 720p, 4K from 1080p. The video keeps its length, measured before anything is charged. For a longer source, the video upscaling guide covers the catalogue's other upscaling models.
Frequently asked questions
- Is Gemini Omni Flash free on Beemm Vision?
- The free plan gives access to the whole catalogue, Gemini Omni Flash included, with 100 free credits. Each generation spends credits according to the version, length and resolution you choose, and the cost is shown before you launch it.
- What is the difference between Gemini Omni Flash 1.2, 1.1 and V1?
- 1.2 is the newest: text, start image, references or a video to edit, 3 to 15 seconds, 360p to 4K, with two companions, Extend and Upscale. 1.1 has the same modes over 3 to 10 seconds. V1, shown as “1.0” in Generate, starts from text, a start image or references, with no last frame, no resolution setting and no video editing.
- How long can a Gemini Omni Flash video be?
- 3 to 15 seconds with 1.2 (10 at most when a last frame is set), 3 to 10 with 1.1 and V1. An edit lasts as long as its source video, 15 seconds at most on 1.2, and Extend then adds 3 to 10 seconds to an existing video.
- Does Gemini Omni Flash output 4K?
- Yes: versions 1.2 and 1.1 output 360p, 720p, 1080p or 4K, at 16:9 or 9:16. V1 has no resolution setting. Upscale also raises an existing video of up to 15 seconds to 720p, 1080p or 4K.
- Does Gemini Omni Flash generate sound?
- Version 1.2 does: the soundtrack is generated with the video, with no option to switch on. For versions 1.1 and V1, the Generate picker shows what each one produces.
- Where is Gemini Omni Flash in Beemm Vision?
- In the video tab of Generate, under Google: the Gemini Omni Flash line offers versions 1.2, 1.1 and 1.0 (V1) and the Frames, Reference, Text, Edit, Extend and Upscale types. On the canvas each version is a video node; that is where reference clips are wired in.
Try Gemini Omni Flash
The button opens Generate with Gemini Omni Flash 1.2 already selected. The full catalogue is available from the free plan.
Keep reading
- LTX on Beemm VisionGenerate video with Lightricks' LTX 2.5: text, image or audio to video, up to 20 s and 4K, then upscale to 8K, cut out the background or reframe it.
- PixVerse on Beemm VisionGenerate video with PixVerse V6 Max and C1: text, image, a transition between two images or references to video, 1 to 15 seconds, up to 1080p.
- Vidu Q4 on Beemm VisionAnimate an image with Vidu Q4 by ShengShu Technology: image-to-video from 3 to 16 seconds, 540p to 4K, in the image's format, in Generate and the canvas.
- AI video generator: every video modelModels of the same kind on one canvas, with the cost shown before each run.
- The full catalogueEvery image and video model available in the canvas.