Deploy and Host onyx on Railway

Onyx is an open-source AI platform that connects to your team's documents, tickets, and messages so it can chat, search, and act with full context of your organization. It combines RAG-based chat with 40+ data connectors (Slack, Confluence, GitHub, Google Drive, and more), custom AI agents, and enterprise auth — deployable as a self-hosted alternative to closed AI assistants.

About Hosting onyx

Onyx is a multi-service application: a FastAPI backend, a Next.js frontend, a Celery-based background worker, and two dedicated ML model servers for embeddings and reranking, backed by Postgres, Redis, OpenSearch, and S3-compatible object storage. This template wires all ten services together on Railway's private network with a single public entry point behind nginx, so the whole stack comes up healthy with one deploy — no manual service linking, connection strings, or Docker Compose required.

Common Use Cases

Internal AI search and chat across company docs, wikis, tickets, and chat history
Self-hosted ChatGPT-style assistant with your own LLM provider and data connectors
Custom AI agents for support, engineering, or research teams that need RAG over private data

Dependencies for onyx Hosting

An API key from an LLM provider (OpenAI, Anthropic, Azure, or any LiteLLM-supported provider), added after first login
At least a Hobby-tier Railway plan — the model servers and OpenSearch need meaningful memory headroom (roughly 10–14GB combined across services)

Deployment Dependencies

Why Deploy onyx 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 onyx on Railway, you are one step closer to supporting a complete full-stack application with minimal burden. Host your servers, databases, AI agents, and more on Railway.