NOFX - AI trading terminal for global markets.

Why deploy nofx render on Render?

NOFX is an AI-powered trading terminal that supports multiple cryptocurrency exchanges, includes a strategy studio, and features a conversational agent for trade setup. It solves the problem of self-hosting a unified trading interface with AI capabilities, allowing developers to manage trades across exchanges through natural language commands.

This template deploys a complete AI trading terminal with the Go API, React UI, and nginx already wired together in a single container, plus a 5 GB persistent disk for SQLite data—no need to configure reverse proxies, static file serving, or volume mounts yourself. Secrets like JWT_SECRET and DATA_ENCRYPTION_KEY are auto-generated by the Blueprint, and RSA keys bootstrap on first boot. One click gives you a running NOFX instance with persistent storage, versus manually coordinating Dockerfiles, disk provisioning, and environment variable generation across services.

Architecture

What you can build

After deploying, you'll have a self-hosted trading terminal where you can connect exchange accounts (Hyperliquid, OKX, and others), configure LLM providers, and create automated trading strategies through a chat-based agent or manual setup. Data persists across restarts on a 5 GB disk, so your traders, strategies, and API keys survive redeploys. Expect roughly $12/month on the Starter plan.

Key features

• Multi-exchange trading support: Connect to Hyperliquid, OKX, and other exchanges with API key configuration managed through the built-in UI.
• Conversational strategy builder: Use the Agent chat interface to describe trading strategies in natural language before wiring up exchange connections.
• Persistent SQLite storage: A 5GB disk mount at /app/data preserves traders, strategies, and encrypted exchange configs across deploys and restarts.
• Pre-built GHCR images: Dockerfile.render pulls official container images from GitHub Container Registry, avoiding slow monorepo builds on deploy.
• Auto-generated security keys: JWT_SECRET, DATA_ENCRYPTION_KEY, and RSA keys are automatically generated at deploy time with no manual secret configuration required.

Use cases

• Solo trader tests LLM-assisted crypto strategies on Hyperliquid without local setup
• Quant engineer prototypes trading bots using natural language agent chat
• Team lead spins up isolated NOFX instances for each developer demo
• Hobbyist runs 24/7 paper trading bots on persistent cloud infrastructure

What's included

Next steps

1. Open your service URL and complete the registration form to create the admin account — You should see the sign-in screen after registration, confirming the single admin user was created successfully
2. Navigate to Config, add an AI model (OpenAI, Anthropic, or DeepSeek), and save — The model should appear in your configured models list and be available for agent conversations
3. Open the Agent tab and describe a simple test trader in natural language — You should see the AI respond with a trader configuration suggestion, confirming LLM connectivity is working

Resources

• NOFX Github