Gemini Omni SDK

Official Node.js / TypeScript SDK for Gemini Omni — the next-generation unified multimodal AI video generation and conversational video editing platform.

🎬 Explore Online Studio: Try the full interactive studio at https://gemini-omni.dev.

✨ Features

• Multimodal Video Generation: Text-to-Video, Image-to-Video, and Video-to-Video rendering.
• Cinematic Quality: Native 720p, 1080p, and 4K output.
• Character & Style Consistency: Persistent identity retention across scenes.
• Conversational Editing: Modify camera angles, lighting, and elements via natural language prompts.
• TypeScript Support: Full type definitions included out of the box.

📦 Installation

npm install gemini-omni
# or
pnpm add gemini-omni
# or
yarn add gemini-omni

🚀 Quick Start

1. Generate Video from Text Prompt

import { GeminiOmniClient } from 'gemini-omni';

const client = new GeminiOmniClient({
  apiKey: process.env.GEMINI_OMNI_API_KEY,
});

async function main() {
  const task = await client.generateVideo({
    prompt: 'Cinematic drone shot of a futuristic cyberpunk city at twilight, neon reflections in rainwater, 4k ultra-detailed',
    aspectRatio: '16:9',
    resolution: '1080p',
  });

  console.log('Task submitted successfully:', task.taskId);
}

main();

2. Image-to-Video with Character Consistency

import { GeminiOmniClient } from 'gemini-omni';

const client = new GeminiOmniClient();

async function createAvatarVideo() {
  const result = await client.generateVideo({
    prompt: 'A gentle smile looking directly into camera, slow natural blinking, warm cinematic lighting',
    referenceImageUrl: 'https://your-domain.com/character.png',
    aspectRatio: '9:16',
  });

  console.log('Video generated:', result);
}

🔗 Links & Resources

• Official Studio & Generator: https://gemini-omni.dev
• Prompt Library: https://gemini-omni.dev/prompts
• Model Specs & Benchmark: https://gemini-omni.dev/gemini-omni-1-1-flash

📄 License

MIT © Gemini Omni