· Updated 9 November 2026 · 01:06 CET
Gemini Omni 1.2 Flash: fifteen seconds, picture and sound
Google’s new video model joins the library with native audio, clips up to 15 seconds, six image references and video editing.

A short scene needs room for something to happen: a glance, a reply, a camera move that arrives somewhere. Gemini Omni 1.2 Flash is now in the SUPERMODELS video library. It brings a little more time to those ideas, with sound built into the shot.
Set the length of the scene
Clips run from 3 to 15 seconds, with duration adjustable one second at a time. The model generates synchronized audio. You can begin with a prompt, animate a still, or use up to six images as references for a subject or style. Those references guide the scene rather than becoming a sequence of literal frames.
For controlled transitions, supply a first and last frame; that mode supports up to 10 seconds. Choose either frame inputs or subject references. An existing video can also be edited, provided it is no longer than 15 seconds. Video edits use the source length and framing, and cannot be combined with image inputs.
Write the sound into the shot
Try this starting point: “A continuous shot inside a quiet flower shop. The camera moves towards a florist wrapping an orange tulip. She says, ‘This one is for you.’ Paper rustles and a bicycle bell rings outside. No music.” Set the duration in the controls, then adjust one part of the brief after reviewing the result.
A useful prompt gives the action somewhere to go. Decide who moves, what the camera follows and what we hear before adding more visual detail. When editing footage, describe the change and spell out the elements you want to preserve.
Draft small, then choose your output
Use 360p while exploring, or select 720p. The 1080p and 4K options are upscaled outputs. Landscape and portrait framing are available, and the platform estimates the cost from your resolution and clip length before you generate.
Open the Gemini Omni 1.2 Flash model page to review its supported controls and pricing. Gemini Omni 1.1 remains in the library for existing workflows.
Capabilities and limits checked against the official Replicate model documentation and input schema.
Cover image prompt
Editorial cover artwork generated with OpenAI image generation. The exact prompt is below.
Illustrate this everyday scene for the SUPERMODELS illustration collection: A small group of neighbors in simple folding chairs, seen from behind at a courtyard film screening. Alternate cobalt and moss silhouettes, ivory projection light and one orange chair; keep the people distinct and the screen blank.
Contemporary SUPERMODELS editorial gouache illustration with clean cut-paper shapes, fine subtle risograph grain, warm ivory, cobalt blue, moss green and burnt orange, tactile matte painted paper, soft directional light, generous negative space and an elegant playful composition. Limit the palette to those colors and charcoal shadows. Use broad simplified shapes and visible dry gouache texture, as in a printed editorial illustration. Avoid photorealism, 3D rendering, glossy materials, ornate fantasy, clutter, outlines, gradients and collage grids. One self-contained full-bleed landscape artwork, exact 16:9 aspect ratio, with the focal subject comfortably inside the middle 75 percent. No text, lettering, logo, frame, borders or watermark.