RoseKhlifa

RoseKhlifa / Image-Studio

Public

开源image2调用图像生成/编辑桌面客户端 · SSE 流式保活,兼容 Cloudflare 524/504 超时截断 · Wails (Go + React/TS) · 数据 100% 本地

15
1
85% credibility
Found May 19, 2026 at 18 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

Image Studio is a free, open-source desktop application that helps everyday people create and edit images using AI. You can describe what you want to see in plain language, or upload an existing photo and ask the AI to transform it. The app includes a full canvas editor where you can mask areas, draw annotations, rotate, flip, and crop your images. Everything you create saves locally to your computer, and your generation history is stored so you can revisit past results. The app works with any AI service that speaks the standard language most image generators use, letting you connect your own account through a simple settings screen.

How It Works

1
💬 You hear about an AI image tool

A friend mentions a free, open-source app for creating and editing images with AI that runs entirely on your own computer.

2
📥 You download and install the app

You grab the installer from GitHub, and within minutes your new Image Studio is ready to open.

3
🔑 You connect your AI account

The app asks for your AI service address and a key — you paste them in, and everything lights up green.

4
🎨 You describe what you want to create

You type a description like 'a cozy coffee shop at sunset' and pick your preferred size and quality settings.

5
You choose how to create
🖼️
Text-to-Image

Describe your vision and watch the AI bring it to life

✏️
Image-to-Image

Upload a photo and tell the AI how to modify it

6
🖌️ You refine your image

Use the canvas tools to mask areas you want changed, draw annotations, rotate, flip, or crop — then send it back to the AI.

7
💾 Your creations save automatically

Every image lands in your personal folder, and your history grows so you can revisit any result anytime.

🎉 You have your image!

Your AI-generated artwork is ready to share, print, or use in your next project — all without leaving your desk.

Sign up to see the full architecture

6 more

Sign Up Free

Star Growth

See how this repo grew from 18 to 15 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is Image-Studio?

Image-Studio is a desktop application for AI image generation and editing, built with Wails (Go backend paired with a React/TypeScript frontend). It wraps the gpt-image-2 model through compatible OpenAI-style APIs, letting you generate images from text prompts or edit existing images with reference photos. The app stores everything locally -- API keys, generated images, and history -- never touching a remote server except to forward your generation requests upstream. It ships with both a full graphical client and a command-line script for scripted workflows.

Why is it gaining traction?

The standout feature is how it handles Cloudflare 524/504 timeouts -- a notorious pain point when AI image generation runs long. Image-Studio uses SSE streaming with automatic keepalive and retries up to three times on timeout errors, so you do not have to babysit failed requests. The canvas is surprisingly polished: you can paint masks for inpainting, annotate regions with arrows and rectangles, compare results side-by-side with a draggable split slider, and batch-generate multiple images in parallel. The app supports two upstream API modes -- the Responses API with SSE for timeout resilience, or the standard Images API for broader compatibility.

Who should use this?

Developers and power users who want a local-first image generation workflow without browser tabs or one-off scripts. If you are already using a GPTCODEX-compatible relay and want a proper desktop tool instead of curl commands or ChatGPT interfaces, this fills that gap. The CLI script also works for CI/CD pipelines generating assets on demand. At version 0.1.0 with 15 stars, it is early-stage -- fine for experimentation and personal use, but production deployment warrants caution.

Verdict

Image-Studio solves a real problem (timeout-prone AI image generation with a polished local UI) and executes it well for what it is. The 0.85% credibility score reflects its nascent maturity: the codebase shows care, but limited stars and thin documentation mean you should vet it yourself before committing to a workflow. Worth trying if you want desktop-level control over gpt-image-2 without vendor lock-in.

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.