Onyx on Railway

Open-source AI search and chat across your company tools, self-hosted on Railway.

Official site · License: MIT (outside ee directories) · 6 Railway templates

Deploy Onyx on RailwayCompare templates

Onyx (formerly Danswer) is an open-source AI platform for enterprise search and chat. It connects to tools such as Slack, Google Drive, Confluence, Jira and GitHub, indexes their content and answers questions with citations using the LLM of your choice.

Teams use it as a private ChatGPT-style assistant over internal documents, and to build agents scoped to document sets for support, sales enablement or engineering. Self-hosting keeps indexed data and chat history on infrastructure you control.

Onyx is MIT licensed outside its ee directories, which fall under the Onyx Enterprise License.

What Onyx can do

  • Connectors for Slack, Google Drive, Confluence, Jira, GitHub and more
  • Hybrid keyword and vector search with cited answers
  • Chat assistant and custom agents
  • Bring your own LLM provider
  • Local embedding model servers
  • Permission sync with source tools

What people use it for

  • Asking questions across company knowledge with cited answers
  • A private internal AI assistant
  • Agents scoped to support or engineering documents
  • Keeping indexed data on your own infrastructure

Videos

Onyx Demo - Open Source Enterprise Search · Onyx
Onyx - The open-source chat UI (Demo) · Onyx
Onyx Local Deployment Guide · Onyx

Running Onyx on Railway

On Railway, the Onyx template deploys ten services: nginx as the front door, the API server, the web app, a background service, two model servers for embeddings, OpenSearch, Postgres, Redis and SeaweedFS for file storage, all pinned to v4.9.0 with generated secrets and healthchecks. It is memory-intensive: the template author plans for roughly 8 to 14 GB of RAM across services, an estimate rather than a measurement. You need an LLM provider, configured in the admin panel.

Onyx templates on Railway

Onyx v4 Enterprise AI SearchMaintained by Tracklayer
Deploys the full Onyx v4 stack for AI-powered enterprise search and chat: API server, web app, background workers, two local embedding model servers, OpenSearch, Postgres, Redis an
4 deployments
Deploy
Onyx by an independent creator
Onyx - Open Source AI Platform - AI Chat with advanced features
24 deployments
Deploy
Onyx Lite by an independent creator
Onyx Lite - Open Source AI Platform - AI Chat with advanced features
2 deployments
Deploy
Onyx by an independent creator
AI chat and search over your company's documents and tools
1 deployments
Deploy
Onyx Lite Chat + Postgres by an independent creator
Onyx Lite self hosted AI chat with Postgres, four services
1 deployments
Deploy
Onyx by an independent creator
Open-source AI chat and enterprise search over your team's documents
0 deployments
Deploy

Tracklayer templates are listed first. Other Onyx templates follow, ordered by total deployments on Railway. How this works.

Deployment numbers as of 2026-11-20.

Questions about Onyx

How do I log in?

Wait for api-server to become healthy, since the first start runs migrations and can take up to 15 minutes. Open the nginx public URL and create an account; the first account becomes the administrator.

What drives the cost?

Memory. The template author estimates 8 to 14 GB across services, with the background service, both model servers and OpenSearch the largest. Hobby is the minimum plan and Pro is recommended.

Do I need an LLM provider?

Yes. Add OpenAI, Anthropic, Azure, Bedrock, Ollama, a LiteLLM gateway or another provider in the admin panel.

How do I restrict sign-ups?

Set VALID_EMAIL_DOMAINS on api-server or configure SSO in the admin panel.

Which licence applies?

Onyx is MIT outside its ee directories, which use the Onyx Enterprise License and stay disabled in this template. This is a community template, not affiliated with Onyx.

Guides

Similar apps

Tracklayer maintains the templates marked as such; Onyx is a trademark of its owners and this site is not affiliated with them.