Deploy ArcReel

Self-hosted AI video workspace: stories to storyboards and short videos

Deploy ArcReel

/var/lib/postgresql/data

/data

Deploy and Host ArcReel on Railway

ArcReel is an open-source, self-hosted AI video workspace. An AI agent turns a novel or script into characters, scenes, props, storyboards and short videos, keeps them consistent across shots, and exports CapCut/JianYing drafts. It works with many image, video, voice and text model providers and tracks what each generation costs.

About Hosting ArcReel

This template runs ArcReel 0.37.0 from the official Docker image, pinned by version and digest, with Railway's PostgreSQL as the database. All projects, generated images and videos, logs and the agent's runtime data live on a volume at /data, so they survive redeploys. The login password is generated at deploy time: read AUTH_PASSWORD in the ArcReel service's Variables tab, then sign in as admin. Model provider API keys go into ArcReel's Settings page, not into Railway variables. Generation calls paid third-party APIs, so you need at least one provider account (for example Gemini, Volcano Ark or an OpenAI-compatible endpoint).

Common Use Cases

  • Turn a novel, story or script into a storyboard and short-form video episodes
  • Produce narrated or drama-style short videos with consistent characters across shots
  • Draft product ads and export the edit to CapCut/JianYing for finishing

Dependencies for ArcReel Hosting

  • PostgreSQL (Railway's PostgreSQL service, included)
  • An API key for at least one AI model provider, added in ArcReel's Settings page

Deployment Dependencies

Implementation Details

The service builds from aalfath/arcreel-railway-template, whose Dockerfile pins arcreel/arcreel:0.37.0 by digest. A small start script points ArcReel's data root and ~/.claude at the volume, switches Railway's DATABASE_URL to the asyncpg driver, and listens on PORT (1241). Railway checks GET /health.

Upstream's Docker Compose gives the container seccomp:unconfined and NET_ADMIN so bubblewrap can sandbox the agent's Bash tool. Railway containers can't have those, so the template uses ArcReel's own fail-closed fallback: the agent's Bash tool may only run ArcReel's bundled skill scripts. Generation, editing and export run through ArcReel's built-in tools as usual.

Updates: when the template repo moves to a new ArcReel version, Railway offers the update in your project. Read the repo's CHANGELOG.md first and back up the volume and database before large upgrades.

Why Deploy ArcReel 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 ArcReel 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.


Template Content

More templates in this category

View Template
Chat Chat
Chat Chat, your own unified chat and search to AI platform.

okisdev
117
View Template
Weaviate
Vector database with hybrid search and an API key-protected API

GitHub Locker
1
View Template
Möbius Personal AI OS
Self-hosted AI agent that builds and runs apps for you, with memory.

Tracklayer
1