LM Studio Setup Guide

System Requirements

• Windows 10/11 64-bit, macOS 10.15+, or Linux
• Minimum 16GB RAM (32GB recommended)
• GPU with at least 8GB VRAM for optimal performance
• At least 10GB free disk space
• Stable internet connection

Installation Steps

1. Download LM Studio from https://lmstudio.ai/
2. Install the application following your OS-specific instructions
3. Launch LM Studio
4. Download your preferred model

Step 1: Download a Model

1. In LM Studio, click on "Discover" in the sidebar
2. Browse available models - recommended starting models:
   • TheBloke/Mistral-7B-Instruct-v0.2-GGUF - Good balance of performance and size
   • TheBloke/Llama-2-7B-Chat-GGUF - Reliable general-purpose model
   • TheBloke/Gemma-7B-it-GGUF - Google's latest model
3. Click "Download" next to your chosen model

4. Select the quantization level:
   • Q4_K_M - Best balance of speed and quality
   • Q5_K_M - Higher quality but slower
   • Q3_K_M - Faster but lower quality

Step 2: Configure the Model

1. After download, the model will appear in your "Models" list
2. Click on the model to select it
3. To find the exact model name for HelpLink:
   • Click on the Developer tab (second icon from top in the left sidebar - looks like a code/terminal icon)
   • Look at the "llm" field in the interface - this is your model name
   • For example: llama-3.2-3b-instruct
   • Copy this exact name - it's case sensitive!
4. Click "Load Model" to activate it

Step 1: Configure LM Studio Server

1. In LM Studio, click the "Developer" tab (second icon from top)
2. Click the "Settings" button (gear icon)
3. Configure the following settings:
   • Server Port: 1234 (default)
   • Enable CORS: Turn ON (green toggle)
   • Serve on Local Network: Turn ON (orange/yellow toggle)
   • Verbose Logging: Optional - ON for debugging
   • Log Prompts and Responses: Optional - ON for debugging
   • Just-in-Time Model Loading: Optional - ON for memory optimization
4. Click the "Running" toggle to start the server
5. Verify the server is running by checking the "Reachable at:" URL

Step 2: Network Configuration

Find Your IP Addresses:

Configure Port Forwarding:

1. Access your router's admin panel (typically http://192.168.0.1 or http://192.168.1.1)
2. Find Port Forwarding settings
3. Add new rule:
   • Internal Port: 1234
   • External Port: 1234
   • Protocol: TCP
   • Internal IP: Your computer's local IP

Step 3: Configure Firewall

Windows:

1. Open Windows Defender Firewall with Advanced Security
2. Add new Inbound Rule for port 1234

macOS:

Linux:

Final Setup

1. In HelpLink's AI Model Settings:
2. Click "Add Custom Model"
3. Enter your model details:
   • Name: Your choice (e.g., "My Remote LM Studio")
   • Base URL: http://YOUR_EXTERNAL_IP:1234/v1
   • API Key: Just write "no key"
   • Model Name: Copy from LM Studio's Developer tab
4. Click "Add Model" to save
5. Test the connection