Skip to content

Frona is an extremely lightweight application with a small footprint. The hardware you need depends on how you plan to use it. If you only use chat and web search, minimal resources are enough. If you run browser automation with many concurrent sessions, compile code, or deploy multiple apps, you'll need more compute resources.

Hardware

Recommended resources:

ResourceMinimumRecommended
CPU2 cores4+ cores
RAM4 GB8+ GB
Disk20 GB50+ GB SSD
NetworkBroadbandLow-latency connection

Browserless (headless Chrome) is the most resource-intensive component. Each concurrent browser session uses roughly 200-500 MB of RAM. The default configuration allows 10 concurrent sessions.

Software

  • Docker 24 or later
  • Docker Compose v2
  • Linux (recommended for production. Landlock sandboxing only works on Linux)
    • macOS works for development but CLI sandboxing is limited

Network

The engine needs outbound access to:

  • Your LLM provider's API (Anthropic, OpenAI, etc.)
  • The internet (for web search and browsing tools)
  • Twilio API (if using voice features)

Inbound, you need to expose:

  • Port 3001 (Frona serves both the API and frontend on a single port)
  • A public URL if using Twilio voice callbacks

LLM provider

You need an API key from at least one supported provider. Frona supports 17 providers:

Anthropic (Claude), OpenAI (GPT), Groq, OpenRouter, DeepSeek, Google Gemini, Cohere, Mistral, Perplexity, Together AI, xAI (Grok), Hyperbolic, Moonshot, Mira, Galadriel, Hugging Face, and Ollama (self-hosted, no API key needed).

See Environment Variables for the full list of API key variables.

Optional services

  • Twilio account: required only for voice call features
  • ngrok: useful for exposing your local instance to the internet (e.g., for Twilio callbacks during development)