The AI model to use for error explanation.

<p><strong>Recommended models by provider:</strong></p>
<ul>
<li><strong>OpenAI</strong>: 
  <ul>
    <li>gpt-4, gpt-4-turbo, gpt-3.5-turbo, etc.</li>
    <li>More advanced models available https://platform.openai.com/docs/models/</li>
  </ul>
</li>
<li><strong>Google Gemini</strong>:
  <ul>
    <li>gemini-2.0-flash, gemini-2.0-flash-lite, gemini-2.5-flash, etc.</li>
    <li>More advanced models available https://ai.google.dev/gemini-api/docs/models</li>
  </ul>
</li>
<li><strong>Ollama</strong>:
  <ul>
    <li>gemma3:1b, gpt-oss, deepseek-r1, etc. any model available in your Ollama instance</li>
    <li>More advanced models available https://ollama.com/search</li>
  </ul>
</li>
</ul>

<p>The model you choose affects both the quality of explanations and the API cost. Generally, more advanced models provide better explanations but cost more per request.</p>