aitu  by ljquan

AI creative suite for image and video generation

Created 6 months ago
424 stars

Top 69.7% on SourcePulse

GitHubView on GitHub
Project Summary

Opentu (opentu.ai) is an open-source, AI-powered creative suite that integrates AI image and video generation with a versatile visual workspace. It targets creators, engineers, and power users seeking a unified platform for ideation, design, and media production, offering a free and extensible solution to streamline complex creative workflows.

How It Works

Opentu is built upon the Plait framework, a robust drawing engine, and features a plugin-based architecture for modularity and extensibility. The core application provides an infinite canvas supporting freehand drawing, image insertion, and structured diagramming (mind maps from Markdown, flowcharts from Mermaid). It integrates multiple AI models for image generation (e.g., Gemini) and video generation (e.g., Veo3, Sora-2), managed via an asynchronous task queue with features like batch processing and model switching. The architecture supports various UI frameworks and rich text editors, enabling a layered development approach.

Quick Start & Requirements

  • Online Experience: Access production at opentu.ai or the beta at pr.opentu.ai.
  • One-Click Deploy: Options available for Vercel and Netlify.
  • Local Development: Requires Node.js >= 16.0.0 and npm >= 8.0.0. Clone the repository, run npm install, and start with npm start (accessible at http://localhost:7200).
  • Docker: Pull the latest image (ljquan/aitu:latest) and run with docker run -d -p 8080:80 ljquan/aitu:latest (accessible at http://localhost:8080).

Highlighted Details

  • Advanced AI Media Generation: Supports multiple AI models for generating high-resolution images (HD/2K/4K) and videos, with features like reference images, batch generation, and task management.
  • Integrated Visual Canvas: Offers an infinite canvas with freehand drawing, mind mapping, flowchart creation, image insertion, and comprehensive editing tools (undo/redo, copy/paste, multi-select).
  • Extensible Plugin Architecture: Designed with a modular plugin system allowing for custom functionality and integration with different front-end frameworks.
  • Open Source & Free: Licensed under the permissive MIT license, enabling commercial use and modification.

Maintenance & Community

The project actively uses GitHub Discussions for community engagement and GitHub Issues for bug reporting and feature requests. A roadmap details completed, in-progress, and planned features, indicating ongoing development.

Licensing & Compatibility

The project is released under the MIT License, which is highly permissive and allows for commercial use, modification, and distribution without significant restrictions.

Limitations & Caveats

Key collaborative features, such as multi-user real-time editing, PDF/SVG export, a template system, a plugin marketplace, cloud synchronization, and an API platform, are currently in development or planned for future releases. Users working with very large canvas files may experience rendering slowdowns, with a recommendation to segment large projects.

Health Check
Last Commit

3 days ago

Responsiveness

Inactive

Pull Requests (30d)
13
Issues (30d)
10
Star History
223 stars in the last 30 days

Explore Similar Projects

Starred by Peter Norvig Peter Norvig(Author of "Artificial Intelligence: A Modern Approach"; Research Director at Google).

NanoBananaEditor by markfulton

1.1%
626
Advanced AI image generation and editing platform
Created 6 months ago
Updated 5 months ago
Feedback? Help us improve.