For a one-off video clip, use Gemini Omni Flash.

Why Gemini Omni Flash

The reasons it wins for this job.

It ranks first on the text-to-video with audio arena (Elo 1233) and first on image-to-video without audio (1369), as of 17 September 2026.
It generates sound with the picture, so a short promo does not need a separate audio pass.
It is in the Gemini app and Google Flow on Google AI subscriptions, so there is no API to set up for one clip.
Image-to-video lets you start from a product shot or brand still, which keeps the first frame on-brand.

How to use it

4 steps to a first result.

1Start from a stillMake or pick the first frame as an image. Animating a good still beats describing a scene from nothing.
2Describe motion and soundCamera move, what moves in frame, and the audio. One idea per clip.
3Generate several, keep oneVideo models vary a lot between runs. Make four and pick the best.
4Cut it together elsewhereClips are short. Assemble, caption and add your logo in an editor.

Prompt for Gemini or Flow

Start from this.

Edit the parts in capitals, then run it.

Animate the attached image into a LENGTH-second clip for PLACEMENT, aspect ratio RATIO.

Camera: slow push in toward the product.
Motion: SUBTLE MOVEMENT IN FRAME, e.g. steam rising, screen lights up.
Audio: soft ambient room tone and one clear click as the screen turns on. No music, no voice.
Style: keep the lighting and colours of the source image exactly. No text on screen.

Nothing is sent anywhere. It copies to your clipboard.

Alternatives that also work

If Gemini Omni Flash is not an option.

Wan 3.0Official page →

Alibaba · closed · cost: medium

Pick it when you need a longer single clip. Wan 3.0 generates up to 30 seconds and ranks first on the arena for video without audio.

MiniMax H3Hugging Face →

MiniMax · open weights · cost: medium · 4.6M HF downloads/mo

Pick it when you are comfortable with a developer playground. H3 Max on fal ranks within six points of first.

Veo 3.1Official page →

Google · closed · cost: high

Pick it when you already use Google Flow and want Google's longer-standing video model.

Watch out for

OpenAI is shutting down Sora: the app closed in April 2026 and the API is removed on 24 September 2026. Do not start new work on it.
Video arenas change monthly, and several top models launched in the last eight weeks. Check the arena before a big project.

Doing this again and again? For video at volume, generate it with MiniMax H3 on fal. Ad variants and product clips at scale need near-top quality, a simple per-second price and fast generation. H3 Max on fal ranks third on the audio arena, within six Elo points of first, at a lower per-second price than the leader.

Sources

Checked 17 September 2026. Models change monthly; we re-check this page when they do.

Picking the model is the easy part.

Wiring it into a workflow that runs every week, with evals, fallbacks and a cost you can predict, is the work. Fifteen minutes, no deck.

Book fifteen minutes →