Run Open WebUI as a Shared Team Interface

What is Open WebUI?

Open WebUI is a self-hosted web interface for Ollama that looks and feels like ChatGPT. When deployed on a server, your whole team can access it through a browser — no software to install on their machines.

Step 1. Prerequisites

1. 1.A server with Docker installed.
2. 2.Enough RAM for your chosen model (8GB minimum).
3. 3.Ollama installed and running (see our Ollama guide).
4. 4.Check Ollama is working: run ollama list to see installed models.

Step 2. Install Open WebUI

Pull and run the container.

Each flag explained:

• •-d runs in background.
• •-p 3000:8080 maps port 3000 on host to 8080 in container.
• •--add-host lets the container reach Ollama on the host.
• •-v persists data.
• •--restart always ensures it comes back after reboots.

Step 3. Connect to Ollama

1. 1.Open http://server-ip:3000 in your browser.
2. 2.Create an admin account (first user becomes admin).
3. 3.Go to Settings → Models and set the Ollama endpoint to http://host.docker.internal:11434.
4. 4.Your models from Ollama should appear in the dropdown.

Step 4. Set up user accounts

1. 1.Go to Settings → Admin → Users.
2. 2.Create accounts for each team member.
3. 3.Assign roles: admin (full control) or user (chat only).
4. 4.Each user signs in with their own credentials.

Step 5. Create shared knowledge bases

Upload documents (PDFs, text files) to a shared workspace. All team members can query these documents through the chat interface. This turns Open WebUI into a team knowledge assistant.

Step 6. Access controls and monitoring

• •Restrict which models each user can access.
• •Monitor usage logs.
• •Set rate limits if needed.
• •For external access, use a reverse proxy with HTTPS or Tailscale for secure remote access.