Your cross-platform desktop AI application for Windows, Mac, and Linux. Provides powerful general AI capabilities such as code generation, document summarization, and intelligent Q&A to comprehensively boost your productivity.

中文 | English | 日本語

Official Website • Documentation • Quick Start • Changelog • 302.AI Platform

Clean and intuitive conversation interface, supporting multi-model switching, file uploads, tool invocation, and more

Conversation list on the left, multi-tab dialogue window on the right, easily manage multiple conversation threads

Built-in 302.AI toolbox, quickly open various AI applications with one click, no need to switch to browser

Independent settings window, supporting data management, Vibe mode, Skills, MCP servers, and other configurations

Support real-time preview of AI-generated front-end code effects, WYSIWYG development experience

• 🤖 Support for OpenAI, Anthropic, Google, and other major AI providers
• 🔄 Flexible model switching and configuration, switch between different models in the same conversation at any time
• 🎛️ Advanced conversation parameter controls (temperature, top-p, token limits, etc.)
• 📊 MCP (Model Context Protocol) server integration

• 🤖 Integrated Claude Code, support natural language requirement description, AI automatically completes development
• ☁️ Cloud sandbox environment, pre-installed with Node.js, Python, Git, CMake, and other development toolchains, zero configuration out of the box
• 🔒 Independently isolated cloud execution environment, AI operations do not affect local files
• 📋 Task panel displays AI execution process in real-time, achieving batch management and automatic execution of multiple AI tasks
• 🚀 One-click project deployment, instant launch, persistent hosting
• 🎭 Side-by-side preview of code execution results, supporting preview, file tree, and terminal three view modes
• 🧠 Support Plan mode, allowing AI to plan the implementation approach before execution, suitable for complex tasks and architectural design

• 📦 Support 4 creation methods: manual writing, file upload, GitHub import, history generation
• 🔧 Built-in 17 official Skills, ready to use out of the box
• 📝 Visual management interface, easily edit and organize Skills

• 🖼️ Upload images for AI-assisted content analysis and description generation
• 📄 Support for multiple file formats
• 💻 Code syntax highlighting
• 📊 Mermaid diagram visualization
• 📝 Full Markdown rendering support

• 🖥️ Multi-platform support for Windows, Mac, and Linux
• 🌙 Customizable light/dark theme system, support for custom CSS styles
• 👤 Support account login, and query balance and usage
• 📱 Responsive design, perfectly adapts to various screen sizes

• 🗂️ Manage multiple conversation threads simultaneously, clear thinking without confusion
• ⚡ Support for real-time streaming responses
• ⌨️ Complete keyboard shortcut system
• 🧰 Built-in 302.AI tool marketplace, covering 50+ AI application tools

• 📂 Conversation records stored locally, protecting privacy
• ☁️ Support cloud synchronization, cross-device access
• 🕶️ Conversations support incognito mode, no chat records saved
• 📤 Support import and export of conversation history

• 🇨🇳 Chinese
• 🇺🇸 English
• 🇯🇵 Japanese (coming soon)

• Skills Store: Added Skills Store in the client, supporting direct browsing, one-click installation, and immediate use of skills within the client
• Vibe Mode Notification Optimization: When Vibe mode is inactive, task completion will be notified via desktop notifications
• Task Board: Support for configuring task loop count, added AI intelligent decomposition feature
• Help Documentation Entry: Added "View Help Documentation" button for quick access to usage instructions
• Plan Mode: Added Plan mode (planning mode) for better task and plan management
• Web Deployment Management: Support viewing all deployed web pages and deleting deployed web pages directly from the list

• Vibe Mode Model Settings: Support configuring default model for Vibe mode
• Mode Switching Interaction Optimization: Optimized the switching interaction experience between normal chat mode and Vibe mode

• Fixed the issue where editing a Skill name would incorrectly create a new Skill
• Fixed the issue where in Plan mode, after multi-selecting tasks, selection could not be canceled in some cases
• Fixed the issue where uploading txt attachments in Vibe mode could not be correctly synchronized to the sandbox environment

• Vibe Mode: Added Plan mode
• Settings: Support viewing and deleting all deployed web pages
• Task Panel: Added task loop count setting

• Vibe Mode: Support setting default model to use

• Fixed the issue where closing the Vibe mode settings page would mistakenly close the preview window
• Fixed the issue where starting a chat after writing on the task board would clear the task board content

• Task Panel: Added task panel module, supporting task orchestration, automatic execution, and other features
• Tool Extensions: Added Nano-Banana-MD, Nano-Banana-PPT, and 3D Studio tools to the tool marketplace

• Message Copy Optimization: Chat messages support right-click menu to copy selected text content
• Vibe Mode Enhancement: Support viewing associated files during AI conversation

• Fixed the display abnormality issue where conversation content overflowed the chat container boundary
• Fixed the issue where kimi-for-coding model could not be called normally in Vibe mode

• Support displaying changelog

• Claude Skills System: Brand new visual management panel
• Support 4 Skill creation methods (manual/upload/GitHub/history)
• Built-in 17 official Skills, ready to use out of the box

• Operating System: Windows 10+ / macOS 10.14+ / Linux (Ubuntu 18.04+)
• Node.js: 18.x or higher
• Package Manager: pnpm 10.18.3+ (required)
• Memory: 4GB RAM (8GB+ recommended)
• Storage: 500MB available space
• Network: Stable internet connection (to access AI provider APIs)

# 1️⃣ Clone the project
git clone https://github.com/302ai/302-AI-Studio.git
cd 302-AI-Studio

# 2️⃣ Install dependencies
pnpm install

# 3️⃣ Start the development server 🎉
pnpm dev

# Start development server (with hot reload)
pnpm dev

# Type checking
pnpm check

# Code linting
pnpm lint

# Auto-fix linting issues
pnpm lint:fix

# Format code
pnpm format

# Check code formatting
pnpm format:check

# Complete quality check
pnpm quality

# Auto-fix all issues
pnpm quality:fix

# Run unit tests
pnpm test:unit

# Run E2E tests
pnpm test:e2e

# Run all tests
pnpm test

# Build SvelteKit application
pnpm build

# Package Electron app (output in /out directory)
pnpm package

# Create distributable installer
pnpm make

# Publish to configured targets
pnpm publish

📦 302-AI-Studio-sv
├── 📂 src/                          # Renderer process source code
│   ├── 📂 lib/                       # Shared libraries
│   │   ├── 📂 components/            # UI components
│   │   │   ├── ui/                   # Shadcn-Svelte base components (40+)
│   │   │   └── buss/                 # Business components
│   │   │       ├── chat/             # Chat interface
│   │   │       ├── model-*/          # Model selection and configuration
│   │   │       ├── provider-*/       # AI provider management
│   │   │       ├── theme-*/          # Theme system
│   │   │       └── settings/         # Application settings
│   │   ├── 📂 stores/                # State management (Svelte 5 Runes)
│   │   ├── 📂 types/                 # TypeScript type definitions
│   │   ├── 📂 api/                   # API integration layer
│   │   ├── 📂 utils/                 # Utility functions
│   │   ├── 📂 theme/                 # Theme system
│   │   ├── 📂 datas/                 # Static data
│   │   └── 📂 hooks/                 # Svelte Hooks
│   ├── 📂 routes/                    # Routes
│   │   ├── (with-sidebar)/           # Main application layout
│   │   │   └── chat/                 # Chat interface routes
│   │   ├── (settings-page)/          # Settings page layout
│   │   │   └── settings/             # Settings route groups
│   │   └── shell/                     # Shell window routes
│   ├── 📂 shared/                    # Cross-process shared code
│   │   ├── storage/                  # Persistent storage
│   │   └── types/                    # Shared types
│   ├── 📂 messages/                  # Internationalization message files
│   └── 📄 app.html                   # HTML template
├── 📂 electron/                      # Electron main process
│   ├── main/                         # Main process code
│   │   ├── services/                 # IPC services
│   │   ├── generated/                # Auto-generated IPC bindings
│   │   └── constants/                # Electron constants
│   └── preload/                      # Preload scripts
├── 📂 vite-plugins/                  # Custom Vite plugins
│   └── ipc-service-generator/        # IPC service generator
├── 📂 scripts/                       # Build scripts
├── 📂 docs/                          # Documentation
├── 📂 e2e/                           # Playwright E2E tests
└── 📄 package.json                   # Project configuration

We welcome all forms of contributions! Whether it's reporting bugs, suggesting new features, or submitting code improvements.

1. Code Contributions: Submit PRs to improve the code
2. Bug Fixes: Submit fixes for issues you've discovered
3. Feature Suggestions: Have a great idea? We'd love to hear your suggestions
4. Documentation: Help us improve documentation and usage guides
5. Promotion: Spread the word about 302 AI Studio

# 1. Fork the project
# 2. Create a feature branch
git checkout -b feature/amazing-feature

# 3. Commit changes (following Conventional Commits)
git commit -m 'feat: add amazing feature'

# 4. Push to the branch
git push origin feature/amazing-feature

# 5. Create a Pull Request

This project is open source under AGPL-3.0, you are free to use, modify, and distribute it.

302.AI is a pay-as-you-go AI application platform that solves the last-mile problem of applying AI in practice.

1. 🧠 Comprehensive collection of the latest and most complete AI capabilities and brands, including but not limited to language models, image models, audio models, and video models
2. 🚀 Deep application development based on foundation models, developing real AI products rather than simple chatbots
3. 💰 Zero monthly fees, all features are pay-as-you-go, fully open, truly low barriers with high ceilings
4. 🛠️ Powerful management backend, targeting teams and small-to-medium enterprises, one person manages, multiple people use
5. 🔗 All AI capabilities provide API access, all tools are open source and customizable (in progress)
6. 💡 Strong development team, launching 2-3 new applications weekly, with daily product updates. Developers interested in joining are welcome to contact us