AI Image Edit is a web-based platform for AI-driven image generation and localized editing, allowing users to modify specific image regions while maintaining the original style. It targets users who require precise control over AI-generated visuals, offering a solution for refining details without regenerating entire images.

How It Works

This project utilizes a React frontend with Fabric.js as its canvas engine, supporting AI models compatible with the OpenAI API format, including Gemini. Its core functionality revolves around precise inpainting and outpainting, enabling users to select arbitrary areas via brushes or lasso tools and provide specific editing instructions. A multi-layer system and an infinite canvas facilitate complex image composition and manipulation workflows.

Quick Start & Requirements

• Installation: Clone the repository (git clone https://github.com/chunxiuxiamo/ai-image-edit.git). For local development, install dependencies with npm install and run npm run dev (access at http://localhost:5173). Docker deployment is available via docker-compose up -d --build (access at http://localhost:8890).
• Prerequisites: A Node.js environment is required for local development. Docker is necessary for containerized deployment. While not explicitly stated as a setup requirement, practical use of AI models implies the need for suitable hardware (e.g., GPU).
• Links: Project repository: https://github.com/chunxiuxiamo/ai-image-edit.

Highlighted Details

• Intelligent Generation & Editing: Supports text-to-image generation, continuous dialogue, and precise regional editing that preserves the original image style.
• Advanced Selection & Canvas: Features multiple lasso selection tools (Free, Polygonal, Magnetic), AI background removal (@imgly/background-removal), and an infinite canvas with a robust multi-layer system.
• Model Compatibility: Integrates with models supporting the OpenAI API format and offers native Gemini interface support.
• Workflow Enhancements: Includes batch text-to-image generation, reference image support for generation and editing, and comprehensive layer management (copy/paste, reordering, visibility, locking).

Maintenance & Community

Recent updates in January and February 2026 indicate active development. Community interaction primarily occurs through GitHub Issues. No dedicated community channels (e.g., Discord, Slack) or details on major contributors or sponsorships are provided.

Licensing & Compatibility

The project is licensed under the MIT License. This permits commercial use, modification, distribution, and private use, provided that the original copyright and license notices are retained.

Limitations & Caveats

Editing complex images may result in unintended modifications to non-selected areas. Certain AI models, such as Gemini-2.5-flash-image, necessitate explicit mask creation for editing operations. Performance benchmarks and specific hardware requirements for AI model inference are not detailed in the README.