LLM Configuration

Before any exam can be AI-graded, you must configure an LLM provider. This is a one-time setup step.

1. In the sidebar, click LLM Configuration.
2. You will see a form with the following fields:
   • Provider — Choose from:
     • OpenAI
     • Anthropic
     • Google Gemini
     • OpenAI-compatible (DeepSeek, Qwen, Together, OpenRouter, Groq, Ollama, and others)
   • Model — Enter the exact model identifier:
     • OpenAI: gpt-4o, gpt-4-turbo, gpt-3.5-turbo
     • Anthropic: claude-3-5-sonnet-20241022, claude-3-opus-20240229
     • Google: gemini-1.5-pro, gemini-1.5-flash
     • OpenAI-compatible: Check your provider’s documentation (e.g., deepseek-chat, qwen-max)
   • API Key — Paste your API key from the provider’s developer console.
   • Base URL — Only required for OpenAI-compatible providers. Leave blank for OpenAI, Anthropic, and Google. For custom providers, enter the API endpoint (e.g., <your-provider-api-endpoint>).
3. Click Save.
4. Click Test Connection to verify the API key and model work.

Warning: Only one LLM configuration can be active at a time. Saving a new configuration automatically deactivates the previous one. Changes are audited — each field change is logged with the Super Admin who made it.

Tip: If you are using an OpenAI-compatible provider and get a connection error, double-check that the Base URL does not end with a trailing slash and includes the correct protocol.