Gemini Omni Flash, now 1.2: AI video with native audio

Meet Gemini Omni 1.2 Flash, the newest version, with clips and edits up to 15 seconds. Google DeepMind’s model edits your footage with a prompt and generates AI video with sound effects, ambience and lip-synced dialogue built in. Start from text, an image, start and end frames or reference clips.

Generate Now

Who gets the most out of Gemini Omni Flash?

Gemini Omni Flash is built for video creators, educators, and marketers who need to produce and edit high-quality video fast, without a full production crew or complex software.

  • AI tools for social video creators

    Social video creators and YouTubers

    Remix existing footage, swap backgrounds, and stay visually consistent across every shot, just by describing the change. Ideal for YouTube Shorts and YouTube Create.

  • AI tools for educators

    Educators and content explainers

    One prompt can produce a claymation, or animated explainer grounded in real-world science.

  • Avatar creators for marketing and social teams

    Digital marketing teams

    Product stays consistent. Everything else is changeable. Swap the background, or setting to generate ad variants without a reshoot or extra talent.

Get the drone shot without the drone

Gemini Omni Flash drone shot effect turns a still photo into a cinematic aerial clip, with the camera pulling up and away from your subject like a real drone pass. This model keeps fast motion smooth where most models break apart, holding steady through orbits, dives, and fly-throughs. Generate your AI drone video from one image in minutes.

drone shot

What working with Gemini Omni Flash feels like

Gemini Omni Flash changes the editing dynamic more than it changes the output. The latest version, 1.2, brings 15-second generations and edits, up from 10 in 1.1.

Start Creating

  • Clips and edits up to 15 seconds

    Generate or edit 15-second clips, up from 10 in 1.1. Describe the sequence beat by beat, what happens first, next and last, so the extra time is not filled with drift.

  • Your footage is the starting point

    Drop in phone footage, a reference image or a start and end frame. Combine up to 5 images and 3 videos in one generation. Audio references are not supported, but sound is generated with the video.

  • Control the start and the end

    Give a first and last frame and the model generates the motion between them. Keep the two frames close in framing and lighting, since large jumps produce unstable motion.

  • Native audio, up to 4K

    Sound effects, ambience and lip-synced dialogue come out in the same pass as the video, at 360p up to 4K. Lock the style with a 5-second clip at 720p, then raise duration and resolution.

  • Extend a clip, pass by pass

    Extend an existing video by 3 to 10 seconds per pass, at 360p up to 4K. Extension runs as its own toolkit option, Gemini Omni 1.2 Flash Extend.

Gemini Omni 1.2 Flash

Gemini Omni 1.2 Flash generates and edits video with native audio, now up to 15 seconds per clip, from 360p to 4K in 16:9 or 9:16. Add up to 5 images and 3 videos as references, set start and end frames, or extend a clip by 3 to 10 seconds per pass. It is also included in Artlist’s Unlimited plan.

Try Gemini Omni 1.2
Gemini Omni 1.1 flash
Gemini Omni 1.1 flash

Gemini Omni 1.1 Flash

Gemini Omni 1.1 Flash is a multimodal video model that generates and extends clips at resolutions from 360p to 4K. It supports clip extension for videos up to 30 seconds, reference-guided generation with up to 3 videos and 5 images, and explicit first-and-last-frame interpolation for precise motion control.

Try Gemini Omni 1.1

Learn how to create with Gemini Omni Flash

Explore tutorials and creative use cases for AI video generation. Learn how to turn prompts, images, and footage into polished clips using Gemini Omni and Artlist’s AI Tools.

Frequently asked questions

Gemini Omni Flash is Google DeepMind’s AI video generator and editing model. It creates clips with native audio from text-to-video, image-to-video, start and end frames or reference clips, and edits existing footage with a prompt. The latest version is 1.2, and Gemini Omni 1.1 Flash is still available.

The main change is length. Gemini Omni 1.2 Flash generates and edits clips up to 15 seconds, up from 10. Resolutions from 360p to 4K and the 16:9 and 9:16 ratios stay the same as in 1.1.

Yes. Gemini Omni 1.2 Flash is included in Artlist’s Unlimited plan. Unlimited applies to specific models and settings, so check the pricing page for the plans and settings it covers. For anything outside Unlimited, you can switch back to credits.

Gemini Omni Flash is developed by Google DeepMind, Google’s AI research division. It is Google’s Omni video model, which generates clips with native audio from text, images and video. Audio is an output, not an input.

Gemini Omni Flash is first and foremost a video editor, so it fits projects built on existing footage: change the lighting, swap an object or change the camera angle while the core shot stays intact. It also generates from text, images or references.

No. Gemini Omni 1.2 Flash generates native audio with the video, but audio references are not supported. Use up to 5 images and 3 videos as references instead, and describe the sound you want in the prompt.

Edits with Gemini Omni 1.2 Flash can add, remove or replace objects, or change style and lighting, while the core shot stays intact. State both what to change and what to preserve in the prompt, and label reference assets as @Image1 or @Video1. For 15-second clips, describe the sequence beat by beat.

Visit the Artlist blog and Help Center for tutorials on creating with AI video and Artlist’s AI tools. Before any model reaches Artlist, it goes through hands-on evaluation by our team, so you’re working with tools we’ve already vetted.

Still have questions? We're here to help.