Spaces:
Running
Running
File size: 4,924 Bytes
7104219 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 | # Quick Start
## Prerequisites
- Docker Desktop or Docker Engine
- Node.js LTS (for local development only)
- Git
- Modern browser with WebGPU support (for optimal AI performance)
## Installation & Running
### Development (Recommended for Contributing)
```bash
# Clone the repository
git clone https://github.com/felladrin/MiniSearch.git
cd MiniSearch
# Start all services (SearXNG, Node.js app with HMR)
docker compose up
```
Access the application at `http://localhost:7861` (HMR) or `http://localhost:7860` (main)
The development server includes:
- Hot Module Replacement (HMR) on port 7861 for instant code changes
- Full dev tools and source maps
- Live code watching with volume mounts
- Integrated SearXNG search service
- Biome formatting and linting on save
### Production
```bash
# Build and start production containers
docker compose -f docker-compose.production.yml up --build
```
Access at `http://localhost:7860`
Production mode:
- Pre-built optimized assets
- No dev tools or HMR
- Optimized Docker layer caching
## Container Internals
The Docker container runs three services concurrently:
- **SearXNG** - Privacy-focused metasearch engine (starts in background)
- **ONNX Runtime** - In-process inference for result reranking
- **Node.js (Vite)** - Main application server
### Startup Sequence
1. SearXNG webapp launches in background from `/usr/local/searxng/searxng-src`
2. Node.js server starts (dev: `npm run dev` with HMR; production: `npm start -- --host`)
Development mode overrides the container command to run `npm install && npm run dev` before starting.
### Port Configuration
| Port | Service | Variable | Notes |
|------|---------|----------|-------|
| 7860 | Node.js (Vite) | `PORT` | Main application (both dev and production) |
| 7861 | HMR | `HMR_PORT` | Vite HMR server (development only) |
| 8888 | SearXNG | - | Internal metasearch engine (not exposed) |
## First Configuration
### No Configuration Required (Default)
MiniSearch works out of the box with browser-based AI inference. Search works immediately, and AI responses use on-device models via Wllama (WebGPU-accelerated when available, CPU otherwise).
### Optional: Enable AI Response
1. Open the application
2. Click **Settings** (gear icon)
3. Toggle **Enable AI Response**
4. Choose inference type:
- **Browser** (recommended): Uses Wllama for local processing (WebGPU-accelerated when available)
- **Internal API**: Configure OpenAI-compatible endpoint
- **AI Horde**: Pre-configured distributed computing
5. The app will download ~300MB-1GB model files on first use (browser mode)
6. Subsequent loads are instant (cached in IndexedDB)
### Optional: Restrict Access
Add access keys to prevent unauthorized usage:
```bash
# Create .env file
echo 'ACCESS_KEYS="my-secret-key-1,my-secret-key-2"' > .env
# Restart containers
docker compose up --build
```
Users will be prompted to enter an access key before using the app.
## Development Without Docker
```bash
# Install dependencies
npm install
# Start development server
npm run dev
# In another terminal, start SearXNG (or use standalone instance)
# See SearXNG documentation for setup
```
Access at `http://localhost:7860`
**Note**: Running without Docker requires manual SearXNG setup and environment configuration.
## Verification
### Test Search
1. Enter any query in the search box
2. Press Enter or click Search
3. Results should appear within 2-5 seconds
### Test AI Response
1. Toggle "Enable AI Response" in Settings
2. Search for "What is quantum computing?"
3. After search results load, an AI-generated response should appear with citations
### Test Chat
1. After getting an AI response
2. Type a follow-up question like "Tell me more"
3. The AI should respond using conversation context
## Common Issues
### Issue: Search returns no results
**Solution**: Verify SearXNG is running. Check container logs:
```bash
docker compose logs searxng
```
### Issue: AI response never loads
**Solution**: Check browser console for errors. Common causes:
- WebGPU not supported (app will automatically fall back to Wllama)
- Model download blocked by firewall or CORS
- Insufficient disk space for model caching
- Browser extensions blocking WebAssembly
- Try switching inference types in Settings
### Issue: Access key not working
**Solution**: Ensure `ACCESS_KEYS` is set in `.env` file and containers were rebuilt with `--build` flag.
### Issue: Port already in use
**Solution**: Change ports in `docker-compose.yml`:
```yaml
ports:
- "7862:7861" # Use 7862 instead of 7860 (HMR port)
```
## Next Steps
- **Customize AI**: See `docs/ai-integration.md` for model selection and inference options
- **Configure**: See `docs/configuration.md` for all environment variables and settings
- **Architecture**: See `docs/overview.md` for system design
- **Contributing**: See `docs/pull-requests.md` for development workflow
|