---
title: "Deploy Memory"
description: "Deploy and Host Memory with Railway"
category: "AI/ML"
url: https://railway.com/deploy/memory
---

# Deploy Memory

Deploy and Host Memory with Railway

**[Deploy Memory on Railway](https://railway.com/template/memory)**

Machine-readable deploy manifest (JSON, validated by TemplateCI): https://railway.com/deploy/memory/manifest.json

- **Creator:** Maciej Kucharz's Projects
- **Category:** AI/ML

## Template content

### xberg

- **Image:** ghcr.io/xberg-io/xberg:1.3.0
- **Health check:** /health

### postgres

- **Image:** pgvector/pgvector:pg18-trixie

### zitadel

- **Image:** ghcr.io/zitadel/zitadel:v4.17.3
- **Start command:** `/app/zitadel start-from-init --masterkeyFromEnv`
- **Public domain:** Yes

### memory-web-ui

- **Image:** ghcr.io/emergent-company/memory-web-ui:0.96.0
- **Health check:** /api/health
- **Public domain:** Yes

### memory-server

- **Image:** ghcr.io/emergent-company/memory-server:0.96.0
- **Health check:** /ready
- **Public domain:** Yes

## Buckets

- **memory-documents**

## Documentation

# Deploy and Host Memory with Railway

Memory is a self-hosted knowledge graph and memory service for AI agents and teams. This template provisions the full stack with Railway.

## About Hosting Memory

The template creates five services and a Railway bucket: Postgres with pgvector, xberg for document extraction, Zitadel for authentication, the Memory API/MCP server, and the Memory web UI. After deployment, a short one-time Zitadel bootstrap connects the server and web UI.

## Why Deploy Memory

Deploy Memory on Railway to get a managed Postgres database, private service networking, public HTTPS domains, persistent volumes and a reproducible full-stack deployment without running your own Kubernetes cluster.

## Common Use Cases

- Give AI agents persistent memory and a queryable knowledge graph
- Search and organize documents, notes and research
- Run a private MCP server for coding agents
- Host a team knowledge base with authenticated web access

## Dependencies for Memory Hosting

Memory needs Postgres with the pgvector extension, an S3-compatible document store, an OIDC provider, and the Memory server and web UI images.

### Deployment Dependencies

- `pgvector/pgvector:pg18-trixie`
- `ghcr.io/xberg-io/xberg:1.3.0`
- `ghcr.io/zitadel/zitadel:v4.17.3`
- `ghcr.io/emergent-company/memory-server:0.96.0`
- `ghcr.io/emergent-company/memory-web-ui:0.96.0`

Configure your AI provider in Memory after deployment. AI credentials are not baked into the template.


## Similar templates

- [Chat Chat](https://railway.com/deploy/-WWW5r) — Chat Chat, your own unified chat and search to AI platform.
- [Weaviate](https://railway.com/deploy/weaviate-1) — Vector database with hybrid search and an API key-protected API
- [Möbius Personal AI OS](https://railway.com/deploy/mobius-v01-personal-ai-os) — Self-hosted AI agent that builds and runs apps for you, with memory.

Open this page in a browser: https://railway.com/deploy/memory