Open WebUI on Railway
A self-hosted chat interface for language models, with Ollama, search and RAG, on Railway.
Open WebUI is a self-hosted chat interface for language models. It connects to Ollama for local models or to any OpenAI-compatible API, and adds multi-user accounts, document chat with retrieval, web search and tools.
Individuals and teams use it as a private ChatGPT-style front end, so conversations and uploaded documents stay on a server they control while still allowing hosted models through an API key.
Open WebUI uses its own licence based on BSD-3-Clause with an added branding clause: the name and logo must stay visible unless you have 50 or fewer end users in a 30-day period, written permission, or an enterprise licence.
What Open WebUI can do
- Chat interface for Ollama and OpenAI-compatible APIs
- Multi-user accounts with admin approval
- Document chat with retrieval (RAG)
- Web search integration
- Tools and model management in the UI
What people use it for
- A private chat front end for a team
- Running small local models with Ollama
- Chatting with your own documents
- Adding a shared UI in front of an OpenAI-compatible API
Videos
Running Open WebUI on Railway
The Railway template deploys Open WebUI v0.12.0 with Ollama (pulling llama3.2:3b and nomic-embed-text on first boot), SearXNG for web search, Postgres with pgvector for users, chats and vectors, and Redis, with volumes for data. Railway has no GPUs, so models run on CPU and should stay small; memory and storage from Ollama and the models are the main cost drivers.
Open WebUI templates on Railway
Tracklayer templates are listed first. Other Open WebUI templates follow, ordered by total deployments on Railway. How this works.
Deployment numbers as of 2026-11-20.
Questions about Open WebUI
How do I log in?
Enter your email in WEBUI_ADMIN_EMAIL at deploy, then copy WEBUI_ADMIN_PASSWORD from the Open WebUI service variables and sign in at the public URL. New sign-ups stay pending until approved.
Why is the first start slow?
Ollama downloads about 2.3 GB of models on first boot. Wait for its logs to show both models pulled before starting a chat.
Can it run large models?
Not well. Railway has no GPUs, so models run on CPU. For larger models, add an OpenAI-compatible API key.
Are there licence terms to know?
Yes. The licence has a branding clause that keeps the name and logo visible unless you meet one of its exceptions. Read the upstream licence before rebranding.
How do I upgrade?
Images are pinned. Back up Postgres and the data volume, change the Open WebUI tag and redeploy; migrations run on start.
Guides
Similar apps
Tracklayer maintains the templates marked as such; Open WebUI is a trademark of its owners and this site is not affiliated with them.