Deploy and Host Lobe Chat on Railway

Lobe Chat (LobeHub) is an open-source AI chat workspace: one interface for OpenAI, Anthropic, Gemini, Ollama and dozens of other model providers, with knowledge-base file uploads, agents, and multi-user accounts. This template deploys the full server-database edition — the app, a ParadeDB Postgres, Redis, and S3-compatible object storage — so chats, accounts, and uploaded files are stored on your own volumes.

About Hosting Lobe Chat

The template runs four services. lobe-chat is the application, listening on Railway's injected port. postgres is ParadeDB (Postgres 17 with the pg_search and pgvector extensions Lobe Chat requires for its knowledge base), with data on a persistent volume mounted at /var/lib/postgresql. redis holds sessions and caches on a volume at /data. minio is an S3-compatible object store on a volume at /data that keeps file uploads; its bucket is created at boot. The authentication secret, the key-vault encryption secret, the database password, and the storage credentials are all generated at deploy time. Open the public URL, create the first account on the sign-up page, then add your model provider API keys under Settings.

Why Deploy Lobe Chat 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 Lobe Chat 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.

Complete stack — database, cache, and file storage are included and wired together, not left for you to assemble.
Chats and files survive redeploys — every stateful service has its own volume.
Secrets generated at boot — nothing to fill in before the first deploy.
Your keys, your data — conversations and API keys stay in your own project.

Common Use Cases

Private ChatGPT alternative — one self-hosted chat UI for every model provider your team uses.
Team AI workspace — multi-user accounts with shared agents and knowledge bases.
Document Q&A — upload files and chat with them using your own embedding and model keys.

Dependencies for Lobe Chat Hosting

A Postgres database with pg_search (ParadeDB, included) and a volume at /var/lib/postgresql.
Redis (included) and S3-compatible storage (included).
API keys for at least one model provider, added after first login.

Deployment Dependencies

Lobe Chat — upstream project
Lobe Chat documentation — configuration and providers
ParadeDB — the Postgres distribution used here