This quick start guide will get you up and running with OpenLLM Monitor in just a few minutes.
- โ Node.js 18+ installed
- โ MongoDB running (local or cloud)
- โ Git installed
# Clone the repository
git clone <your-repo-url>
cd openllm-monitor
# Install backend dependencies
cd backend
npm install
# Install frontend dependencies
cd ../frontend
npm installCreate backend/.env file:
# Database (Required)
MONGODB_URI=mongodb://localhost:27017/openllm-monitor
# Server Configuration
PORT=3001
NODE_ENV=development
FRONTEND_URL=http://localhost:5173
# LLM Provider API Keys (Add the ones you use)
OPENAI_API_KEY=your-openai-key-here
OPENROUTER_API_KEY=your-openrouter-key-here
MISTRAL_API_KEY=your-mistral-key-here
# Ollama (if using local LLMs)
OLLAMA_BASE_URL=http://localhost:11434Open 3 terminals:
Terminal 1 - MongoDB (if using local MongoDB):
# Start MongoDB service
mongodTerminal 2 - Backend:
cd backend
npm run devTerminal 3 - Frontend:
cd frontend
npm run dev- Open your browser
- Go to:
http://localhost:5173 - You should see the OpenLLM Monitor dashboard! ๐
- Navigate to Logs page
- The system automatically logs all LLM requests
- Check the Dashboard for real-time metrics
Open a new terminal and test the API:
# Test health endpoint
curl http://localhost:3001/api/health
# Test logs endpoint
curl http://localhost:3001/api/logsdocker run -d -p 27017:27017 --name mongodb mongo:latest# Windows PowerShell
.\scripts\setup-mongodb.ps1
# Windows Command Prompt
.\scripts\setup-mongodb.bat- Go to MongoDB Atlas
- Create a free cluster
- Get connection string
- Update
MONGODB_URIin.env
โ
Backend: Visit http://localhost:3001/api/health - should return {"status":"OK"}
โ
Frontend: Visit http://localhost:5173 - should show the dashboard
โ Database: Check the logs show connection to MongoDB
โ WebSocket: Dashboard should show "Connected" status
After setup, verify everything is working correctly:
- Frontend accessible at http://localhost:3000
- Backend API responding at http://localhost:3001/api/health
- MongoDB connected (check logs)
- WebSocket connection established
- Dashboard loads without errors
- Real-time updates working
- Navigation between pages works
- No console errors in browser
- Can view existing logs
- Can create new test requests
- Analytics data displays correctly
- Provider configurations accessible
- MongoDB collections created
- Sample data visible
- Queries performing well
- Indexes created properly
# Test backend health
curl http://localhost:3001/api/health
# Test API info
curl http://localhost:3001/api/info
# Test logs endpoint
curl http://localhost:3001/api/logs
# Test provider endpoint
curl http://localhost:3001/api/providersA healthy API should return:
{
"status": "healthy",
"timestamp": "2024-01-01T00:00:00.000Z",
"version": "1.0.0",
"database": "connected"
}- Go to Settings โ Providers
- Add your API keys
- Test connections
- Your LLM requests will automatically appear in Logs
- View real-time metrics on Dashboard
- Analyze performance in Analytics
- Go to Replay page
- Test prompts across different providers
- Compare responses and costs
- Make sure MongoDB is running
- Check the connection string in
.env - Try:
mongod --versionto verify installation
- Check if backend is running on port 3001
- Verify CORS settings in backend
- Check browser console for errors
- Verify API keys are correct
- Check if you have credits/quota
- Test API keys directly with provider APIs
- Backend (3001): Kill the process or change PORT in
.env - Frontend (5173): Vite will automatically try the next available port
- ๐ Read the full User Guide
- ๐ง Check API Documentation
- ๐ Review Deployment Guide
- ๐ Report issues on GitHub
You now have OpenLLM Monitor running! Start making LLM requests and watch them appear in real-time on your dashboard.
Next Step: Read the complete User Guide to learn about all the powerful features available.