⚙️ Agent Architecture Settings
Configure LLM inference, search engine plugins, browser depth, and Ritual verification.
OpenAI (GPT-4o)
Best overall synthesis
Google Gemini
Fast & multimodal
Claude (Anthropic)
Superior reasoning
Groq (Llama 3.1)
Ultra low-latency
OpenRouter
Multi-model gateway
Ollama (Local)
Self-hosted LLM
Keys are stored client-side in localStorage and never logged.