gpt_image_playground
支持 GPT Image、Gemini(Nano Banana)、Grok 和 Seedream 5.5 的图片生成与编辑工具
What is gpt_image_playground?
支持 GPT Image、Gemini(Nano Banana)、Grok 和 Seedream 5.5 的图片生成与编辑工具
Overview of gpt_image_playground
What it is
GPT Image Playground is an open-source image generation and editing tool that supports multiple AI models including GPT Image, Gemini (Nano Banana), Grok, and Seedream 5.5. It provides a Web UI and a Windows client, allowing users to generate images from text, edit with reference images, and manage history locally. The tool supports official APIs, OpenAI-compatible APIs, sub2api, fal.ai, and custom vendor configurations for non-standard APIs.
Key features
- Supports multiple models and vendors: GPT Image, Gemini, Grok, and Seedream 5.5, with the ability to save multiple configurations and quickly search and switch models.
- Allows integration of non-standard APIs through custom vendor configuration of request parameters and result parsing, without modifying code.
- Supports batch generation with real-time preview, streaming preview, progress tracking, and failure retry.
- Offers Agent multi-turn dialogue for continuous image generation and editing, referencing history, switching branches, and searching the web.
- Provides a local gallery with search, filtering, multiple favorites, batch download, and backup.
- Works on desktop and mobile with support for paste, drag-and-drop, box selection, and mobile swipe multi-select.
Use cases
- Generating images from text prompts using multiple AI models.
- Editing images with reference images, local masks, and canvas annotations.
- Batch generating multiple images or prompts for efficient content creation.
- Using Agent mode for iterative image creation and modification through conversation.
- Managing and organizing generated images and history locally.
Getting started
Windows users can download the desktop client from Releases. For web deployment, see the repository README for instructions on deploying to Vercel, GitHub Pages, Cloudflare Workers, Docker, or a static file server. bash # Docker deployment example docker run -d -p 8080:80 -e DEFAULT_API_URL=https://api.openai.com/v1 ghcr.io/cooksleep/gpt_image_playground:latest
License and pricing
The project is licensed under MIT. Pricing is not stated; it is open-source and free to use.
Based on the project's public README and repository metadata.