Openjourney provides an open-source, high-fidelity clone of the MidJourney web interface, enabling AI-powered image and video generation. It targets developers and power users seeking a familiar UI to leverage Google's Gemini SDK, specifically Imagen 4 for images and Veo 2/3 for video. The project offers a benefit of an extensible platform for AI media creation, allowing integration of custom models alongside Google's offerings.

How It Works

The project is built using a modern Next.js 15 stack with the App Router and Turbopack, styled with Tailwind CSS v4, and animated with Framer Motion. It integrates Google's GenAI SDK to power its core generation capabilities: Imagen 4 for generating a 4-image grid, Veo 3 for text-to-video, and Veo 2 for image-to-video conversions. This approach provides a polished, interactive user experience that mirrors MidJourney's workflow while utilizing powerful, state-of-the-art AI models.

Quick Start & Requirements

• Primary install: Clone the repository, navigate to openjourney-app, and run npm install.
• Prerequisites: Node.js 18+ (20+ recommended), npm or yarn, and a Google AI API Key obtained from Google AI Studio.
• Development Server: Set up a .env.local file with your GOOGLE_AI_API_KEY and run npm run dev. The application will be accessible at http://localhost:3000.
• Links: Google AI Studio: https://aistudio.google.com/

Highlighted Details

• AI Image Generation: Utilizes Imagen 4 to produce four high-quality 1024x1024 images per prompt, presented in a 4-image grid layout.
• AI Video Generation: Supports Veo 3 for text-to-video generation (producing ~3-5 second, 720p videos with audio) and Veo 2 for animating static images based on prompts.
• Interactive UI: Features real-time generation with loading animations, hover effects for video autoplay and interaction, a film strip for navigation, and a lightbox for viewing media.
• Tech Stack: Next.js 15, Turbopack, TypeScript, Tailwind CSS v4, Framer Motion, ShadCN UI, Google GenAI SDK, Radix UI.

Maintenance & Community

The project follows a standard GitHub contribution model (fork, branch, pull request). It acknowledges MidJourney for UI inspiration, Google Gemini API for generation models, Vercel for deployment, and ShadCN for UI components. Further AI projects from the author can be found via @ammaar on X.

Licensing & Compatibility

This project is licensed under the MIT License. This permissive license allows for commercial use, modification, and distribution, including integration within closed-source applications.

Limitations & Caveats

The project relies on external Google AI APIs, which may be subject to rate limits, usage costs beyond the free tier, or policy changes. As a UI clone, its core AI capabilities are dependent on the underlying Google models rather than being self-contained open-source AI models. The README does not specify alpha/beta status, but it is a relatively new project.