AI Integration

Torrify features a flexible AI system that integrates Large Language Models (LLMs) to assist with coding and design.

Supported Providers

1. Google Gemini (BYOK)

• Best For: Fast, free/cheap usage.
• Requirement: Google AI Studio API Key.
• Model: Defaults to gemini-3-flash.

2. OpenRouter (BYOK)

• Best For: Accessing many models (GPT-4, Claude 3, Llama 3).
• Requirement: OpenRouter API Key.

3. Ollama (Local)

• Best For: Privacy, offline use, no API costs.
• Requirement: Ollama running locally.
• Default Endpoint: http://127.0.0.1:11434.

4. PRO (Managed)

• Best For: Convenience.
• Requirement: Torrify PRO License Key.

Configuration

1. Open Settings (⚙️) > AI Configuration.
2. Enable AI Assistant: Toggle ON.
3. Access Mode:
   • PRO: Enter your License Key.
   • BYOK: Select a Provider (Gemini/OpenRouter/Ollama) and enter your API Key.
4. Model: Select or type the model name.
5. Save.

Features

Context Awareness

The AI automatically receives:

• Current Code: The content of your active editor tab.
• Chat History: Previous messages in the conversation.
• Knowledge Base: API documentation for the active CAD backend (OpenSCAD/build123d).

Streaming

Responses stream in real-time. You can stop a generation at any time using the Stop button.

Image Import

Attach images to your chat (paperclip icon) to ask for help modeling from reference photos.

Troubleshooting

• "API key not configured": Check Settings > AI Configuration.
• "Failed to connect": Verify your internet connection or local Ollama server status.
• Generic/Wrong Code: Try being more specific or ask the AI to "check the knowledge base" if it's using outdated syntax.