600aaa937c
- Refreshed the list of available free models to match the application's source code. - Added new models like Qwen3 Next, Gemma 3, and Mistral Large 3.
262 lines
9.2 KiB
Markdown
262 lines
9.2 KiB
Markdown
# 🚀 OpenQode v1.01 - Goose Ultra Edition
|
|
|
|
**The Next-Generation AI-Powered IDE for Modern Developers.**
|
|
*Powered by Qwen AI • Built with Electron • Full-Stack Development Made Easy*
|
|
|
|
|
|
|
|
---
|
|
|
|
## 🚀 Full-Auto Installation (One Command)
|
|
|
|
Run this in your terminal to clone, install dependencies, and build everything automatically:
|
|
|
|
**Windows (PowerShell):**
|
|
```powershell
|
|
git clone https://github.com/roman-ryzenadvanced/OpenQode-Public-Alpha.git; cd OpenQode-Public-Alpha; npm install --legacy-peer-deps; cd bin/goose-ultra-final; npm install --legacy-peer-deps; npm run build; cd ../..; Write-Host "✅ OpenQode is ready. Run OpenQode.bat to start." -ForegroundColor Green
|
|
```
|
|
|
|
**macOS / Linux (Bash):**
|
|
```bash
|
|
git clone https://github.com/roman-ryzenadvanced/OpenQode-Public-Alpha.git; cd OpenQode-Public-Alpha; npm install --legacy-peer-deps; cd bin/goose-ultra-final; npm install --legacy-peer-deps; npm run build; cd ../..; echo -e "\033[0;32m✅ OpenQode is ready. Run ./OpenQode.sh to start.\033[0m"
|
|
```
|
|
|
|
---
|
|
|
|
## 🎯 What is OpenQode?
|
|
|
|
OpenQode is a comprehensive AI coding assistant that brings the power of large language models directly to your development workflow. The flagship **Goose Ultra** application provides a beautiful, feature-rich IDE experience.
|
|
|
|
---
|
|
|
|
## ⚡ Quick Start (Manual)
|
|
|
|
### Windows
|
|
```batch
|
|
# Run the installer (first time only)
|
|
Install.bat
|
|
|
|
# Launch OpenQode
|
|
OpenQode.bat
|
|
```
|
|
|
|
### macOS
|
|
```bash
|
|
# Make scripts executable and install
|
|
chmod +x install-macos.sh OpenQode.sh
|
|
./install-macos.sh
|
|
|
|
# Launch OpenQode
|
|
./OpenQode.sh
|
|
```
|
|
|
|
### Linux (Ubuntu/Debian/Fedora/Arch)
|
|
```bash
|
|
# Make scripts executable and install
|
|
chmod +x install-linux.sh OpenQode.sh
|
|
./install-linux.sh
|
|
|
|
# Launch OpenQode
|
|
./OpenQode.sh
|
|
```
|
|
|
|
---
|
|
|
|
## 📋 Launch Menu Options
|
|
|
|
| Option | Name | Description |
|
|
|--------|------|-------------|
|
|
| **1** | 🚀 **GOOSE ULTRA** | The flagship IDE experience (Recommended!) |
|
|
| **2** | Goose Ultra DEV | Development mode with hot-reload |
|
|
| **3** | Next-Gen TUI (Gen 5) | Terminal-based interface with Ink |
|
|
| **4** | TUI Classic (Gen 4) | Lightweight terminal interface |
|
|
| **5** | Qwen Authentication | Login/refresh Qwen credentials |
|
|
| **8** | Smart Repair | Fix common issues automatically |
|
|
|
|
---
|
|
|
|
## ✨ Features
|
|
|
|
### 🖥️ Goose Ultra IDE
|
|
|
|
The main attraction - a full-featured desktop IDE built with Electron:
|
|
|
|
|
|
|
|
- **Visual Blueprint System** - Describe what you want, get a plan, approve, and build
|
|
- **Live Preview** - See your application rendered in real-time
|
|
- **Code Editor** - Monaco editor with syntax highlighting
|
|
- **Multi-Persona Chat** - Switch between different AI personalities
|
|
- **File Attachments** - Drag and drop files for context
|
|
- **Project History** - All your projects saved and accessible
|
|
|
|
### 🔧 Vi Control Panel (Enhanced)
|
|
|
|
Advanced automation capabilities with new robust connectivity:
|
|
|
|
- **Computer Use** - AI-controlled desktop automation
|
|
- **Browser Control** - Automated web interactions
|
|
- **Vision Analysis** - Screenshot and analyze UI elements with Qwen VL
|
|
- **Remote Hosts** - Enhanced SSH connection management with automatic password fallback
|
|
- **Credential Vault** - Secure password/key storage
|
|
- **Local Engine** - Quick Diagnostics and PowerShell integration
|
|
|
|
### 🎨 Design System
|
|
|
|
- **Apex Mode** - Elite-level code quality enforcement
|
|
- **Competitive Intelligence** - AI researches top competitors for design inspiration
|
|
- **Mobile-First** - All generated code is responsive by default
|
|
|
|
### 📦 UX Package Generator (New!)
|
|
|
|
Export your entire frontend (HTML/CSS/JS) into a single, portable `ux_package.json` file.
|
|
|
|
- **Portable Payload** - Perfect for loading into other AI coding agents (Gemini, ChatGPT, Claude) to build the backend logic.
|
|
- **One-Click Export** - Instantly package your project's frontend artifacts and instructions.
|
|
- **Backend-Ready** - Includes specific prompts instructing the AI how to implement the server-side logic based on your frontend.
|
|
|
|
---
|
|
|
|
## 🛠️ System Requirements
|
|
|
|
### Minimum
|
|
- **Node.js** 18.0 or higher
|
|
- **npm** 8.0 or higher
|
|
- **4GB RAM** for TUI mode
|
|
- **8GB RAM** for Goose Ultra IDE
|
|
|
|
### Recommended
|
|
- **Node.js** 20.x LTS
|
|
- **16GB RAM** for best performance
|
|
- **SSD storage** for fast project loading
|
|
|
|
---
|
|
|
|
## 📁 Project Structure
|
|
|
|
```
|
|
OpenQode/
|
|
├── bin/
|
|
│ ├── goose-ultra-final/ # Main Electron IDE
|
|
│ ├── opencode-ink.mjs # Gen 5 TUI
|
|
│ ├── opencode-tui.cjs # Gen 4 TUI
|
|
│ └── smart-repair.mjs # Auto-repair tool
|
|
├── Documentation/ # All docs and planning files
|
|
├── assets/ # Screenshots and images
|
|
├── Install.bat # Windows installer
|
|
├── install-macos.sh # macOS installer
|
|
├── install-linux.sh # Linux installer
|
|
├── OpenQode.bat # Windows launcher
|
|
├── OpenQode.sh # macOS/Linux launcher
|
|
└── README.md # This file
|
|
```
|
|
|
|
---
|
|
|
|
## 🔐 AI Models & Authentication
|
|
|
|
OpenQode supports **multiple AI providers** for maximum flexibility:
|
|
|
|
### 🟢 Qwen Cloud (Default - Free)
|
|
|
|
Alibaba's powerful AI models with free tier access:
|
|
|
|
1. **First Launch** - The app will prompt you to authenticate
|
|
2. **Browser Login** - A browser window opens for Qwen login
|
|
3. **Token Storage** - Credentials are stored locally and encrypted
|
|
4. **Auto-Refresh** - Tokens are refreshed automatically
|
|
|
|
Available Qwen models: `qwen-coder-plus`, `qwen-plus`, `qwen-turbo`
|
|
|
|
If you need to re-authenticate, use **Option 5** in the launcher menu.
|
|
|
|
---
|
|
|
|
### 🔵 Ollama Cloud (Free - 20+ Models)
|
|
|
|
Access **state-of-the-art open-weight models** for FREE via Ollama Cloud:
|
|
|
|
#### 🎁 Get Your Free API Key:
|
|
1. Go to **[ollama.com/settings/keys](https://ollama.com/settings/keys)**
|
|
2. Sign in or create a free account
|
|
3. Generate an API key
|
|
4. Paste it in Goose Ultra's **AI Model Manager** (sidebar → AI Models)
|
|
|
|
#### 🚀 Available Free Models:
|
|
|
|
| Model | Size | Category | Best For |
|
|
|-------|------|----------|----------|
|
|
| **GPT-OSS 120B** | 120B | Flagship | OpenAI's open-weight reasoning model |
|
|
| **DeepSeek V3.2** | MoE | Flagship | Superior reasoning & agent performance |
|
|
| **DeepSeek V3.1** | 671B | Flagship | Hybrid thinking/non-thinking mode |
|
|
| **Gemini 3 Pro Preview** | Cloud | Flagship | Google's most intelligent model with SOTA reasoning |
|
|
| **Gemini 3 Flash Preview** | Cloud | Fast | Frontier intelligence built for speed |
|
|
| **Qwen3 Coder 480B** | 480B | Coding | Agentic coding, long context |
|
|
| **Qwen3 Coder 30B** | 30B | Coding | Agentic coding model |
|
|
| **Devstral 2 123B** | 123B | Coding | Multi-file editing, software agents |
|
|
| **Devstral Small 2 24B** | 24B | Coding | Vision + tools for software engineering agents |
|
|
| **RNJ-1 8B** | 8B | Coding | Essential AI model optimized for code and STEM |
|
|
| **Qwen3 Next 80B** | 80B | Reasoning | Strong parameter efficiency and inference speed |
|
|
| **Kimi K2** | MoE | Reasoning | State-of-the-art coding agent tasks |
|
|
| **Kimi K2 Thinking** | MoE | Reasoning | Moonshot AI's best open-source thinking model |
|
|
| **Cogito 2.1** | 671B | Reasoning | Instruction tuned generative model |
|
|
| **Qwen3 VL 235B** | 235B | Vision | Most powerful vision-language model in Qwen family |
|
|
| **Qwen3 VL 32B** | 32B | Vision | Powerful vision-language understanding |
|
|
| **Gemma 3 27B** | 27B | Vision | Most capable model that runs on a single GPU |
|
|
| **Ministral 3 14B** | 14B | Fast | Edge deployment, fast responses |
|
|
| **Ministral 3 8B** | 8B | Fast | Edge deployment with vision + tools |
|
|
| **Nemotron 3 Nano** | Nano | Fast | Efficient, open, and intelligent agentic model |
|
|
| **GLM 4.6** | Large | Flagship | Advanced agentic, reasoning and coding |
|
|
| **MiniMax M2** | Large | Flagship | High-efficiency LLM for coding agents |
|
|
| **Mistral Large 3** | MoE | Flagship | Multimodal MoE for production-grade tasks |
|
|
|
|
...and more! Open the **AI Model Manager** in Goose Ultra to see the full list.
|
|
|
|
#### 📖 Ollama Cloud Docs:
|
|
- API Documentation: [docs.ollama.com/cloud](https://docs.ollama.com/cloud)
|
|
- Model Library: [ollama.com/search?c=cloud](https://ollama.com/search?c=cloud)
|
|
|
|
---
|
|
|
|
## 🐛 Troubleshooting
|
|
|
|
### "Node.js not found"
|
|
Run the installer script for your platform:
|
|
- Windows: `Install.bat`
|
|
- macOS: `./install-macos.sh`
|
|
- Linux: `./install-linux.sh`
|
|
|
|
### "Build failed"
|
|
```bash
|
|
# In the goose-ultra-final directory:
|
|
cd bin/goose-ultra-final
|
|
npm install --legacy-peer-deps
|
|
npm run build
|
|
```
|
|
|
|
### "Qwen authentication failed"
|
|
Use **Option 5** (Qwen Authentication) in the launcher to re-authenticate.
|
|
|
|
### TUI crashes or hangs
|
|
Use **Option 8** (Smart Repair) to automatically diagnose and fix issues.
|
|
|
|
---
|
|
|
|
## 🔗 Links & Community
|
|
|
|
- **GitHub:** [roman-ryzenadvanced/OpenQode-Public-Alpha](https://github.com/roman-ryzenadvanced/OpenQode-Public-Alpha)
|
|
- **Telegram:** [@openqode](https://t.me/VibeCodePrompterSystem)
|
|
- **Discord:** [Join Community](https://discord.gg/2nnMGB9Jdt)
|
|
|
|
---
|
|
|
|
## 📄 License
|
|
|
|
This project is provided as-is for educational and development purposes.
|
|
See [LICENSE](LICENSE) for details.
|
|
|
|
---
|
|
|
|
*Made with ❤️ by @RomanRyzenAdvanced*
|
|
*Powered by Qwen AI • Developed with [TRAE.AI IDE](https://www.trae.ai/)*
|