🤖 Setup Open WebUI — Self-Hosted LLM Chat Interface
Deploy Open WebUI on Ubuntu — the most polished self-hosted chat interface for Ollama and OpenAI-compatible LLMs with RAG, plugins, and multi-user support.
📦 Resources & Setup Scripts
Grab the automated bash script from GitHub to follow along with the video.
Quick Install:
wget https://raw.githubusercontent.com/mhmdali94/Docker/main/AI/open-webui/open-webui-ubuntu.sh
chmod +x open-webui-ubuntu.sh
sudo bash open-webui-ubuntu.shTutorial Steps
1 Download the Script
wget https://raw.githubusercontent.com/mhmdali94/Docker/main/AI/open-webui/open-webui-ubuntu.sh2 Make it Executable
chmod +x open-webui-ubuntu.sh3 Run the Installer
The script installs Docker if needed, pulls the Open WebUI image, and connects it to your Ollama instance automatically.
sudo bash open-webui-ubuntu.sh4 Access the Web Interface
Open your browser and navigate to:
http://<your-server-ip>:3000Ports Used
| Port | Purpose |
|---|---|
| 3000 | HTTP Web Interface |
Overview
Open WebUI is a feature-rich, extensible self-hosted web interface originally built as a polished frontend for Ollama. It has evolved into a comprehensive AI chat platform supporting OpenAI-compatible APIs, document upload for RAG, a plugin and function system, image generation, voice input, persistent conversation history, multi-user authentication, and collaborative workspaces. It runs as a single Docker container backed by SQLite and requires minimal configuration — just point it at your Ollama instance and it is ready.
Why Use It
Ollama by itself requires the terminal or bare API calls. Open WebUI gives your team the polished ChatGPT experience they expect — conversation history, file uploads for document Q&A, model switching, access control, and a full admin panel — all running locally without cloud dependency or per-token fees. For teams, it converts a developer-only tool into a production AI assistant accessible to everyone.
When You Need It
Who Should Use It
Real Use Cases
Main Features
How to Use After Installation
Security Best Practices
Ports and Firewall Notes
Open WebUI runs on port 3000 by default. This port must not be exposed directly to the internet — reverse proxy it through Nginx Proxy Manager or Traefik with a valid SSL certificate. For team deployments, restrict port 3000 to localhost and only allow access through the HTTPS reverse proxy. Ollama communicates with Open WebUI internally on port 11434 and does not need to be publicly accessible when both run on the same host.
Backup and Maintenance
Common Mistakes
Troubleshooting
Alternatives
LibreChat provides a similar multi-model interface with stronger OpenAI/Anthropic aggregation features. AnythingLLM focuses more deeply on RAG and document knowledge base management for enterprise teams. Flowise offers a visual no-code pipeline builder for more complex AI workflows. Direct Ollama Web UI is built into Ollama itself but is minimal — good for quick testing, not for team use.
When Not to Use It
If you only need quick LLM inference without a UI, the Ollama CLI is simpler. For code assistant use, IDE plugins like Continue integrate Ollama directly in your editor more efficiently. If you require enterprise SSO, audit logging, and compliance features, a managed LLM service may be more appropriate. For single-user lightweight usage, Ollama's built-in minimal UI may be sufficient.
PrismaTechWork Professional Help
PrismaTechWork provides end-to-end infrastructure services — from initial deployment and security hardening to ongoing monitoring, automated backups, and dedicated support.
Frequently Asked Questions
Can Open WebUI work without Ollama?
Yes. Set OPENAI_API_KEY and OPENAI_API_BASE_URL environment variables to point to any OpenAI-compatible API. You can mix Ollama for local models and cloud APIs simultaneously from the same interface.
Does Open WebUI support multiple users?
Yes. Open WebUI has a full user management system with admin, user, and pending roles. New user sign-ups can require admin approval. Each user has private conversation history, and admins can control which models each user can access.
Can I enable image generation in Open WebUI?
Yes. Go to Admin → Settings → Images and configure the AUTOMATIC1111 WebUI API URL, a ComfyUI endpoint, or OpenAI DALL-E API key. Once configured, you can generate images directly in the chat window.
How do I update Open WebUI?
Pull the latest image (docker pull ghcr.io/open-webui/open-webui:main), then stop and remove the container and recreate it with the same run command. Your data volume persists across updates. Check the release notes for migration steps before upgrading.
Does document RAG work with Arabic text?
Open WebUI uses standard text extraction for RAG. Arabic PDFs with proper Unicode encoding work fine. The quality of answers also depends on the underlying LLM's Arabic language capabilities.