Deploy and Host Nanochat on Railway

Nanochat is an open-source AI chat client you host yourself. It connects to every frontier model through the nano-gpt API — Claude, GPT, Gemini, and more — from one clean interface. Your team gets a private ChatGPT-style workspace with no per-seat fees, and your conversations stay on your server.

About Hosting Nanochat

Nanochat runs as a TypeScript app backed by PostgreSQL. Deploy both from this template and you get a web UI your team opens from any browser. It handles image and video generation, web search, deep research, speech-to-text, YouTube transcripts, and MCP tool integration out of the box. You need a nano-gpt.com API key to connect to the models.

Common Use Cases

Teams that want a shared AI chat workspace without paying per seat for ChatGPT or Claude
Developers who switch between models (Claude for writing, GPT for code, Gemini for research) and want one interface
Companies that need conversations stored on their own infrastructure for compliance
Power users who want MCP tool integration, image generation, and web search in one chat

Dependencies for Nanochat Hosting

A nano-gpt.com API key for accessing frontier models (Claude, GPT, Gemini, etc.)
PostgreSQL database (included in this template)
No GPU required — Nanochat calls external model APIs, it doesn't run inference locally

Why Deploy Nanochat on Railway?

Railway is a singular platform to deploy your infrastructure stack. Railway will host your infrastructure so you don't have to deal with configuration, while allowing you to vertically and horizontally scale it.

By deploying Nanochat on Railway, you get a private AI chat workspace that runs 24/7 without managing servers. Two services (app + Postgres), one click to deploy, and Railway handles scaling as your team grows.