--- title: "Deploy Multica AI" description: "Manage AI coding agents like teammates. Assign work, track progress & ship." category: "AI/ML" url: https://railway.com/deploy/multica-ai --- # Deploy Multica AI Manage AI coding agents like teammates. Assign work, track progress & ship. **[Deploy Multica AI on Railway](https://railway.com/template/multica-ai)** - **Creator:** INF Labs - **Category:** AI/ML ## Template content ### multica-backend https://raw.githubusercontent.com/multica-ai/multica/main/docs/assets/logo-dark.svg - **Image:** ghcr.io/multica-ai/multica-backend:latest ### multica-web https://raw.githubusercontent.com/multica-ai/multica/main/docs/assets/logo-dark.svg - **Image:** ghcr.io/multica-ai/multica-web:latest - **Public domain:** Yes ### Postgres https://devicons.railway.app/postgres - **Image:** pgvector/pgvector:pg17 ## Documentation # Deploy and Host Multica on Railway Multica is a self-hosted platform for managing AI coding agents like teammates. It helps you assign work, track progress, coordinate agent workflows, and manage development activity from a centralized interface. This template deploys Multica with its Web UI, Backend API, and PostgreSQL with pgvector in a complete Railway stack. ![Multica](https://multica.ai/_next/image?url=%2Fimages%2Flanding-hero.webp\&w=1920\&q=85\&dpl=dpl_5i77AH733wd5yNwW9ppf4vhYvCgm) ## About Hosting Multica Hosting Multica on Railway gives you a complete self-hosted environment for coordinating AI coding agents and their workflows. This template deploys three services: * **Multica Web** — browser-based user interface * **Multica Backend** — API, authentication, WebSocket, agent coordination, and application logic * **PostgreSQL + pgvector** — persistent relational and vector data storage Multica Web and Backend communicate through Railway networking, while PostgreSQL remains private inside the project. The Backend also uses persistent storage for uploaded files so application data survives redeployments. ## Dependencies for Multica Hosting Multica requires the following services and persistent storage to run as a complete self-hosted stack on Railway. | Component | Purpose | Required | | ------------------------- | ------------------------------------------------------ | :------: | | **Multica Web** | Browser-based user interface | ✅ | | **Multica Backend** | API, authentication, WebSocket, and agent coordination | ✅ | | **PostgreSQL 17** | Primary relational database | ✅ | | **pgvector** | Vector extension used inside PostgreSQL | ✅ | | **PostgreSQL Volume** | Persists database data across redeployments | ✅ | | **Backend Upload Volume** | Persists uploaded application files | ✅ | | Redis | External cache or queue service | ❌ | | S3-compatible storage | Optional alternative for file storage | ❌ | | Separate vector database | Not required because pgvector is included | ❌ | This template packages the required infrastructure into three Railway services: ```text Multica ├── Multica Web ├── Multica Backend └── PostgreSQL + pgvector ``` The Multica Backend connects to PostgreSQL through Railway private networking. Multica Web and Backend expose the endpoints required by browser clients, AI agents, and external integrations. Persistent Railway volumes are used for PostgreSQL and Backend uploads so important application data remains available across redeployments. ## Common Use Cases * Manage AI coding agents from one dashboard * Assign development work to agents * Track tasks and agent progress * Coordinate multiple coding agents * Monitor autonomous development workflows * Manage agent workspaces and development projects * Connect tools such as Claude Code, Codex, Cursor, and other coding agents * Run a self-hosted AI development control plane ## Why Use Multica? | Capability | Multica | | --------------------------------- | :-----: | | Centralized AI agent management | ✅ | | Web-based management UI | ✅ | | Multi-agent workflows | ✅ | | Task assignment and tracking | ✅ | | Real-time updates with WebSockets | ✅ | | PostgreSQL persistence | ✅ | | Vector storage with pgvector | ✅ | | Self-hosted deployment | ✅ | | Persistent file uploads | ✅ | | Private database networking | ✅ | | External Redis required | ❌ | | Separate vector database required | ❌ | | Mandatory S3 storage | ❌ | Multica is useful when individual coding agents are no longer enough and you need a central place to coordinate work across agents, projects, and development workflows. ## Architecture The Railway template uses three independent services: ```text Internet │ ┌───────────┴───────────┐ │ │ ▼ ▼ Multica Web :8080 Multica Backend :8080 PUBLIC PUBLIC │ │ │ Railway Networking │ └───────────┬───────────┘ │ ▼ PostgreSQL :5432 PRIVATE │ ▼ pgvector ``` Persistent storage is attached to the required services: ```text Multica Backend │ ▼ /app/data/uploads PostgreSQL │ ▼ /var/lib/postgresql/data ``` ## Multica Web Multica Web provides the main browser interface. It runs on: ```text 8080 ``` Generate a Railway public domain for this port. Example: ```text https://your-multica.up.railway.app ``` The frontend communicates with the Backend through both private and public endpoints depending on whether the request originates from the server or the user's browser. ## Multica Backend The Multica Backend handles: * Authentication * User and workspace management * Agent coordination * Project management * API requests * WebSocket connections * Database access * File uploads * Agent-related application logic The Backend runs on: ```text 8080 ``` A public domain is required because browser clients, external agents, daemons, and integrations may need to reach the API and WebSocket endpoints. ## PostgreSQL and pgvector Multica uses PostgreSQL as its primary persistent database. This template uses PostgreSQL with the **pgvector** extension so relational and vector data can be stored in the same database service. The database runs internally on: ```text 5432 ``` PostgreSQL does not need a public domain for the default deployment. Applications communicate with it through Railway private networking. ## Persistent Storage ### PostgreSQL A persistent Railway volume is mounted at: ```text /var/lib/postgresql/data ``` This preserves: * User accounts * Workspaces * Projects * Tasks * Agent information * Application metadata * Vector data * Database indexes and schemas ### Backend Uploads A second persistent Railway volume is attached to the Multica Backend at: ```text /app/data/uploads ``` This preserves uploaded files across application redeployments and container replacements. ## Public and Private Networking ### Public Services The following services require public Railway domains: | Service | Port | Public | | --------------- | -----: | :----: | | Multica Web | `8080` | ✅ | | Multica Backend | `8080` | ✅ | | PostgreSQL | `5432` | ❌ | ### Private Communication PostgreSQL stays inside the Railway private network. Conceptually: ```text Multica Backend │ │ Railway Private Network ▼ PostgreSQL :5432 ``` The frontend server can also communicate with the Backend through Railway private networking, while browser requests use the Backend's public URL. ## Multica Stack Benefits | Benefit | Supported | | --------------------------- | :-------: | | Complete Web UI | ✅ | | Backend API included | ✅ | | PostgreSQL included | ✅ | | pgvector included | ✅ | | Persistent uploads | ✅ | | Persistent database | ✅ | | Railway private networking | ✅ | | Public WebSocket access | ✅ | | No Redis dependency | ✅ | | No separate vector database | ✅ | The result is a relatively compact stack despite supporting agent management, authentication, persistent data, vector workloads, real-time communication, and file storage. ## Getting Started After Deployment 1. Deploy the Multica template. 2. Wait for PostgreSQL, Multica Backend, and Multica Web to become healthy. 3. Generate a public domain for **Multica Backend** on port `8080`. 4. Generate a public domain for **Multica Web** on port `8080`. 5. Open the Multica Web URL. 6. Create your first account. 7. Create or configure a workspace. 8. Connect your coding agents and development tools. 9. Start assigning work and monitoring agent progress. PostgreSQL should remain private and does not need a public TCP proxy for normal Multica operation. ## Upgrading Multica This template uses the official Multica container images. After deployment, you can change the deployed Multica version directly from the Railway service settings. To change versions: 1. Open the required Multica service in Railway. 2. Go to **Settings**. 3. Locate the container image configuration. 4. Change the image tag to the version you want to run. 5. Redeploy the service. For production deployments, using pinned image versions instead of `latest` can make upgrades more predictable. Always review upstream release notes before performing major upgrades. ## Why Deploy Multica 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 Multica on Railway, you get the Web UI, Backend API, persistent PostgreSQL with pgvector, private networking, WebSocket connectivity, and persistent file storage in one self-hosted AI agent management stack. ## Similar templates - [Chat Chat](https://railway.com/deploy/-WWW5r) — Chat Chat, your own unified chat and search to AI platform. - [stella](https://railway.com/deploy/stella) — Self-host stella with web, API, Postgres, Redis, and object storage. - [Hermes Agent | OpenClaw Alternative with Dashboard](https://railway.com/deploy/hermes-agent-or-openclaw-alternative-wit) — Self-Hosted Hermes AI Agent for Telegram, Discord & Slack Open this page in a browser: https://railway.com/deploy/multica-ai