Using Cline with VS Code and Our Lab's AI Server

1. Install Cline Extension

  1. Open VS Code
  2. Go to Extensions (Ctrl+Shift+X)
  3. Search for "Cline"
  4. Install the extension by Saoud Rizwan
  5. Restart VS Code if prompted

2. Configure Cline

  1. Click the Cline icon in the left sidebar to open the Cline panel
  2. Click the gear icon (⚙️) at the top of the panel
  3. Set the following:
    • API Provider: Select "OpenAI Compatible"
    • Base URL: https://chat.catholic-u.ai/api/v1
    • API Key: YOUR_API_KEY
    • Model ID: RedHatAI/Qwen3-Coder-Next-NVFP4

To use the other model, change the Model ID to: - Sehyo/Qwen3.5-35B-A3B-NVFP4

3. Using Cline

Chat / Ask Questions

  1. Open the Cline sidebar
  2. Type your question or request in the input box
  3. Cline will respond and can read your project files for context

Agentic Coding

Cline can autonomously: - Read and edit files in your project - Run terminal commands (with your approval) - Create new files - Debug errors by reading output and fixing code

Simply describe what you want, e.g.: - "Add input validation to the login form" - "Fix the bug in main.py line 42" - "Write unit tests for the API endpoints"

Cline will ask for your approval before making changes.

Add Context

  • @file: Type @ and select a file to include as context
  • @folder: Reference an entire folder
  • @url: Paste a URL for Cline to read
  • @problems: Include VS Code's Problems panel output

4. Available Models

Model ID Description
RedHatAI/Qwen3-Coder-Next-NVFP4 Qwen3 Coder Next - optimized for coding tasks
Sehyo/Qwen3.5-35B-A3B-NVFP4 Qwen3.5 35B - general purpose

Switch models by clicking the gear icon and changing the Model ID field.

5. Important Notes

  • Function calling: Cline's agentic features (file editing, terminal commands) require the model to support tool use / function calling. If these features don't work, use Cline in chat-only mode.
  • Approval required: Cline will always ask for your approval before editing files or running commands. You can adjust this in settings.
  • Token usage: Cline displays token usage per message in the chat panel.

6. Troubleshooting

  • "Could not connect" error: Make sure you are connected to the university network or VPN.
  • No response: Verify the Base URL, API Key, and Model ID are correct in settings.
  • Agentic features not working: The model may not support function calling. Cline will still work for chat and code suggestions.
  • Slow responses: The server is shared — response times may vary depending on load.