← 返回公开榜单
ja

11cafe/jaaz

AI design agent, local alternative for Lovart. AI agent with ability to design, edit and generate images, posters, storyboards, etc.

本周新增
4,785 星
近 3 周倍数
—
距上次推送
251 天
距上次发布
455 天
许可证
NOASSERTION

星标增长

每根柱子是这个周期记录下来的 star 增长:优先取逐个 stargazer 到达时间分桶的到达数,还没被 stargazer 扫描覆盖的仓库改用该周期的 star 净增;两个写者都没记录的周期画成虚线空档,而不是当成 0。柱高按各自窗口的峰值归一,两张图不共用坐标轴。指向任意一根柱子, tooltip 里是该周期的九项指标:当前值与周期内的净增。

指向或点击一根柱子,看当天的九项指标。

12/1712/301/111/242/62/193/33/16

2026 年 3 月 16 日 新增 4,785 个星标

窗口内最高的一天是 2026 年 3 月 16 日,新增 4785 个星标。

证据时间轴

  1. 第 2026 年第 12 周发布 8 次

    累计 27 个 release

README
The document stored for this repository, with its Chinese translation when one has been produced

🦄 Jaaz –⭐Open source AI Design Agent⭐

https://jaaz.app/

中文入门指南

The world's first open-source, locally deployed creative AI design tool, supporting image generation and editing, unlimited canvas rapid iteration, can call local models and cloud apis, and intelligently batch generate images in one sentence!

Download for macOS    Download for Windows
Screenshot 2025-06-02 at 3 51 56 AM Screenshot 2025-06-02 at 3 51 56 AM Screenshot 2025-06-02 at 3 03 49 PM

✨ Key Features

Smart Prompt Agent Powered by LLMs, Jaaz can interpret your ideas and generate optimized prompts for high-quality image or storyboard creation.

Hybrid Model Deployment Works with local models via Ollama, ComfyUI, and remote APIs like Replicate, OpenAI, or Claude. Run 100% locally or connect to the cloud.

Simply access all the strongest apis After logging in, you can use all the latest models (gpt-image-1,flux kntext,google……)

Interactive Image Editing Supports advanced operations such as object insertion, style transfer, and multi-character coherence via Flux Kontext — all controllable via chat.

Infinite Canvas & Storyboarding Design layouts, plan scenes, and create visual storytelling flows with drag-and-drop simplicity.

(Coming Soon) Video Agent:Video generation and editing via models like Wan2.1, Kling, and more, seamlessly integrated with your creative workflow. Intelligently invoke custom comfyui workflows


  • Use Claude, OpenAI, Gemini via API key, or run locally with Ollama for 100% free usage
  • Use image generation models like GPT-4O, Recraft, Flux, Google Imagen, etc. through Replicate API key

Usage

  1. Click on the "Log in" button at the top right corner of the program's home page, and then you can use the apis of each model. By paying a low price, you can smoothly use the API models
Screenshot 2025-06-02 at 3 08 51 PM
  1. You can also customize the binding API,Add LLM API key like OpenAI or Claude, or install Ollama to use local models
  2. Add image generations API key like Replicate
Screenshot 2025-06-02 at 3 08 51 PM
  1. Start chatting with agent to generate stories or storyboards!
Screenshot 2025-06-18 at 16 09 05

Cases

  • Prompt: Help me place this character in six different scenes, all in front of landmark buildings from around the world. The lighting is harmonious. He takes photos from all over the world, realistic, with warm light, high picture quality, and a picture ratio of 9:16

814c563b08f6ef44de0c2c31f0fdd00b-min

Screenshot 2025-06-02 at 3 51 56 AM Screenshot 2025-06-02 at 3 03 49 PM

API Providers

Deepseek

To use deepseek as a provider, setup as below

Screenshot 2025-06-10 at 12 05 35

Stay Updated

Star Jaaz on GitHub to receive instant noifications about new version releases.

Screenshot 2025-06-10 at 12 05 35

Manual Install (For Linux or local builds)

🟠 Need Python version >=3.12

First git clone this repo:

git clone https://github.com/11cafe/localart

cd react

npm install --force

npx vite build

cd ../server

pip install -r requirements.txt

python main.py

Development

🟠 Need Python version >=3.12

cd react

npm install --force && npm run dev

cd server

pip install -r requirements.txt

python main.py

页面上的每个数字都来自记录下来的 stargazer 到达时间,没有评分公式,也没有 AI 判定。 任何结论都可以用 /api/v1/repos/{id}/stats 里的原始增量自己重算。