Unlock Your Local LLMs: A Guide to Installing & Setting Up Open WebUI (Formerly Ollama WebUI) – May 23, 2025

Hey AI enthusiasts and privacy advocates! Today, we’re diving into a fantastic tool that puts the power of Large Language Models (LLMs) right onto your local machine: Open WebUI (you might also know it by its former name, Ollama WebUI).

As of May 23, 2025, running LLMs locally is becoming increasingly popular. Why? Control, privacy, cost-savings, and the sheer cool factor of having your own AI powerhouse. Open WebUI provides a sleek, user-friendly interface to interact with models run via Ollama.

Let’s get started!

What is Open WebUI?

Open WebUI is a self-hosted, web-based user interface designed to make interacting with local Large Language Models a breeze. It primarily works with Ollama, a popular tool that simplifies running open-source LLMs like Llama 3, Mistral, Phi, and many others directly on your computer.

Think of it as a private ChatGPT-like experience, but you’re in complete control of the models and your data.

The Big Benefits of Running Open WebUI Locally

Why bother setting this up when you can use cloud-based LLMs? Here are some compelling reasons:

1. Ultimate Privacy & Data Control: Your prompts and conversations stay on your machine. No sending data to third-party servers. This is huge for sensitive information or proprietary work.
2. No API Costs or Rate Limits: Once you have the hardware, running inferences is free. Experiment as much as you want without worrying about token costs or usage limits.
3. Offline Accessibility: Once set up, you can often use your local LLMs even without an internet connection (depending on the model and your setup).
4. Customization & Experimentation: Easily switch between different open-source models, tweak parameters, and explore the cutting edge of LLM technology at your own pace.
5. Learning & Development: It’s a fantastic way to understand how these models work under the hood and integrate them into your own projects.

Prerequisites: What You’ll Need

Before we jump into installation, make sure you have the following:

• Docker: Open WebUI is best installed via Docker. If you don’t have it, download and install Docker Desktop (for Mac/Windows) or Docker Engine (for Linux) from the official Docker website.
• Ollama Installed and Running: Open WebUI needs Ollama to serve the LLMs.
  • Download and install Ollama from ollama.com.
  • Ensure Ollama is running.
• At Least One LLM Downloaded via Ollama: You need a model to chat with! Open your terminal or command prompt and pull a model. For example, to get Llama 3:ollama pull llama3content_copydownloadUse code with caution.Bash(You can replace llama3 with other models like mistral, phi3, etc.)

How to Install Open WebUI (Using Docker)

This is the recommended and easiest method.

Step 1: Open Your Terminal or PowerShell

Step 2: Run the Open WebUI Docker Command

Copy and paste the following command into your terminal and press Enter. This command will download the latest Open WebUI image, run it in a detached container, map port 3000 on your host to port 8080 in the container (Open WebUI’s default internal port), set up a volume to persist data, and ensure it communicates with your Ollama instance.

docker run -d -p 3000:8080 -e OLLAMA_BASE_URL=http://host.docker.internal:11434 -v open-webui:/app/backend/data --name open-webui --restart always ghcr.io/open-webui/open-webui:main

content_copydownloadUse code with caution.Bash

Let’s break down that command:

• docker run -d: Runs the container in detached mode (in the background).
• -p 3000:8080: Maps port 3000 on your computer to port 8080 inside the container. You’ll access Open WebUI via http://localhost:3000. If port 3000 is taken, you can change the first number (e.g., -p 8888:8080).
• -e OLLAMA_BASE_URL=http://host.docker.internal:11434: Crucial for Docker Desktop (Mac/Windows). This tells Open WebUI where to find your Ollama service.
  • For Linux users: If Ollama is running on your host machine (not in Docker), you might need to use http://127.0.0.1:11434 or find your host’s Docker bridge IP (often http://172.17.0.1:11434). Alternatively, for Linux, you can run the Docker container with –network=host which simplifies things but is less isolated:# For Linux users, an alternative with host networking: # docker run -d --network=host -v open-webui:/app/backend/data -e OLLAMA_BASE_URL=http://localhost:11434 --name open-webui --restart always ghcr.io/open-webui/open-webui:main # If using --network=host, you don't need -p, it will use port 8080 by default.content_copydownloadUse code with caution.Bash
• -v open-webui:/app/backend/data: Creates a Docker volume named open-webui to store your Open WebUI data (like user accounts, chat history) so it persists even if you update or restart the container.
• –name open-webui: Gives your container a friendly name.
• –restart always: Ensures the container automatically restarts if it crashes or if you reboot your system.
• ghcr.io/open-webui/open-webui:main: The official Open WebUI Docker image.

(Always check the official Open WebUI GitHub repository for the very latest recommended Docker command, as things can evolve.)

Step 3: Wait for it to Start
Docker will download the image (if it’s your first time) and start the container. This might take a few minutes.

Initial Setup and Using Open WebUI

1. Access Open WebUI: Open your web browser and go to http://localhost:3000 (or whatever host port you mapped).
2. Create an Account: The first time you visit, you’ll likely be prompted to create an admin account. Sign up.
3. Select a Model: Once logged in, you should see an interface similar to other chat AIs. In the chat interface, there’s usually a dropdown or a way to select the model you want to use (e.g., “llama3”, “mistral”) from the ones you’ve pulled with Ollama.
4. Start Chatting! Type your prompts and interact with your locally hosted LLM.

Open WebUI often automatically detects your Ollama instance if it’s running on the default http://localhost:11434. If you have any issues connecting, double-check the OLLAMA_BASE_URL environment variable in your Docker command and ensure Ollama is running and accessible.

Enjoy Your Private AI Playground!

That’s it! You now have a powerful, private, and customizable interface for your local Large Language Models. Experiment with different models, explore its features like RAG (Retrieval Augmented Generation) capabilities if you delve deeper, and enjoy the freedom of self-hosted AI.

What are your favorite local LLMs to run with Open WebUI? Share your experiences and tips in the comments below!