We currently support two LLM options:

Option 1: Gemini (Requires API Key)

1. Download Settings: Click the download icon (↓) in the left navigation pane to export your current settings as a JSON file.
2. Edit Settings: Open the downloaded JSON file in a text editor and configure the LLM settings:
   • Set the type field to "gemini"
   • Add your apiKey from Google AI Studio
   • Configure any other required parameters
3. Upload Settings: Click the upload icon (↑) in the left navigation pane and select your modified JSON file to import the settings.

Note: Your API key will be stored locally in your browser and is never sent to our servers. Only you have access to it.

Option 2: MLC AI Web LLM (Completely Private)

MLC AI Web LLM downloads a language model directly into your browser cache and runs entirely on your device. This option is:

• 100% Private: All processing happens locally on your machine
• No API Key Required: No external services needed
• Uses Your Device: Requires client machine compute power and initial model download

Note: First-time setup will download the model to your browser cache. Performance depends on your device's capabilities.