Gemini Omni APIThe Conversational Video Editor
Edit video by just describing what you want. Gemini Omni is Google DeepMind's conversational video editor on fal.ai. Swap subjects, change scenes, restyle, and refine across multiple turns while everything stays consistent. Bring text, images, audio, or video, with native audio, via a fast serverless API.
Editing, As Easy As a Conversation.
Edit clips with prompting
Modify clips, adjust backgrounds, swap subjects, and change styles with plain-English, multi-turn conversation, with no complex timeline software. Each instruction builds on the last, like Nano Banana for video.
Everything Stays In Place
People, objects, and scenes stay consistent from one edit to the next. Restyle a subject or change the world around them while the things you didn't mention hold steady, with no drift or warping.
Cinematic by default
Direct real camera moves like orbital dollies and hero angles, with film-grade lighting, crisp detail, and synchronized native audio. Every shot comes out looking directed, not generated.
Editing first, generation included
Modify clips, adjust backgrounds, and change styles with plain-English, multi-turn conversation, with no complex timeline software. Or start a new video from text, an image, or reference frames. Pick the endpoint for how you want to begin.
Edits generated video across multiple conversational turns while preserving scene coherence. Applies iterative changes through natural-language instructions without regenerating the full sequence from scratch.
Creates video with synchronized audio from text input. Grounded in Gemini's real-world knowledge, with improved physics understanding for more coherent motion and interaction.
Animates a still image into video with audio. Extends a single frame into coherent motion, grounded in Gemini's physical understanding of how scenes and subjects behave.
Generates video with audio from combined multimodal references. Accepts text, images, audio, and video together as input to guide subject, motion, style, and sound in the output.
See what Gemini Omni can create
Copy any prompt and open it in the playground. Each card shows the endpoint it uses; the ones marked Edit run on the edit endpoint with your own source video.
Sketch a route on a map → real aerial footage
"Here's a screenshot of a map with a route I sketched across it. Generate cinematic drone footage flying that exact path at golden hour, and don't show my drawing in the final shot."
Edit videos with prompts
"Replace the pomegranates with apples."
Restyle a subject, identity intact
"Keep me walking in the same shot and change my outfit every second: streetwear, a red dress, a yellow trench. Same pose, same street, no warping."
Ten cities in a single shot
"One continuous hyperlapse through ten world cities: Tokyo, Paris, New York, Venice, Dubai, each location morphing into the next, snapping on the beat."
Bring a still image to life
"Slowly pull back from the eyes of this Renaissance portrait to reveal the whole painting, then let her breathe and smile as clouds drift in the background."
Studio-grade product shots from a prompt
"Place these headphones on a sunlit marble countertop and orbit slowly as soft morning light moves across the surface. Premium product shot."
How to access the
Gemini Omni API
Access Gemini Omni through the fal API in just a few lines of code. Install the fal client for Python or JavaScript, authenticate with your API key, and submit a request to the google/gemini-omni-flash endpoint. The client handles the full request lifecycle: submitting the job, polling for status updates, and returning the finished video when it's ready.
import { fal } from "@fal-ai/client";
const result = await fal.subscribe("google/gemini-omni-flash/edit", {
input: {
video_url: "https://example.com/source.mp4",
prompt: "Change the background to a snowy mountain at dusk. Keep the subject and everything else the same.",
},
logs: true,
onQueueUpdate: (update) => {
if (update.status === "IN_PROGRESS") {
update.logs.map((log) => log.message).forEach(console.log);
}
},
});
console.log(result.data);
console.log(result.requestId);
Common questions about Gemini Omni
What can I create with Gemini Omni?
Gemini Omni is an any-to-any model: bring text, images, audio, video, or sketches and turn them into video with native audio. Generate from scratch, animate a reference, or refine an existing clip through conversation with the edit endpoint.
How does conversational editing work?
Describe a change in plain language: swap a character, adjust lighting, stabilize the shot, or change the background, and Gemini Omni applies it. Edits build across multiple turns while characters, objects, and scenes stay consistent from cut to cut.
Does Gemini Omni generate audio?
Yes. Clips come with synchronized native audio, including ambient sound, effects, and music, generated alongside the video, so there's no separate audio pass.
How do I get started with the Gemini Omni API?
Install the fal SDK (Python or JavaScript), grab an API key from your dashboard, and make your first request in three lines of code. The API is serverless, so no GPUs to manage and no infrastructure to set up. Check the API documentation for all available parameters.
Can I use Gemini Omni for commercial projects?
Yes. Videos generated through the fal API can be used in commercial projects. Check fal.ai's terms of service for full details on usage rights and licensing.
Get in touch about Gemini Omni
Want to learn more about integrating Gemini Omni into your workflow? Leave your details and our team will reach out.