Easily Install Open WebUI & Use It on Your Windows PC/Computer

If you want to run an AI chat interface locally on your Windows PC, Open WebUI is an interesting option. It provides a browser-based interface that can connect to local AI systems such as Ollama, as well as cloud and other compatible AI providers. This means you can manage conversations, models, files, knowledge, and other AI features from one interface.

Open WebUI is self-hosted, which means you can run it on your own computer rather than depending entirely on a separate hosted website. It can work with local models and can also connect to providers such as OpenAI and Anthropic through compatible APIs.

For Windows users, the easiest approach for this guide is to install Docker Desktop, run the official Open WebUI Docker image, and then access Open WebUI through your web browser.

Table of Contents

Understanding Open WebUI

Before installing Open WebUI, it helps to understand what it actually does. Open WebUI is not itself an AI model. Instead, it provides an interface through which you can interact with AI models and providers.

Chat and Conversations

The main feature is the chat interface. After connecting a model, you can start conversations just like you would with other popular AI chat services. Open WebUI can also allow you to switch between available models while working with conversations.

Local AI Model Support

Open WebUI can connect to Ollama, allowing you to use supported AI models running on your own computer. This can be useful if you want to experiment with local AI without sending every prompt to a cloud service. Open WebUI can also help manage Ollama models through its administration interface.

Cloud AI Provider Support

Open WebUI is not limited to local models. It can connect to OpenAI, Anthropic, OpenAI-compatible APIs, vLLM, llama.cpp, and other providers. This gives you a single interface for different AI backends.

File and Document Interaction

You can also use Open WebUI to work with files. For example, after choosing a suitable model, you can upload a document to a conversation and ask questions about its contents. This makes the interface useful for document-based AI tasks.

Knowledge and RAG

Open WebUI includes features for working with knowledge bases and RAG, allowing AI models to work with your own information and documents. This can be useful when you want responses based on your own reference material rather than only the model’s general knowledge.

Web Search and Tools

Open WebUI also provides features such as web search, tool calling, plugins, and Open Terminal. These features can expand what connected AI models can do, although some require additional configuration.

How to Easily Install Open WebUI & Use It on Your Windows PC or Computer

To install Open WebUI on a Windows PC, install Docker Desktop with WSL 2, start Docker Desktop, and run the official Open WebUI Docker command. Docker downloads and starts the Open WebUI container, after which you can open http://localhost:3000 in your browser. Create the administrator account, connect an AI provider such as Ollama, choose a model, and start chatting. Open WebUI stores its application data in a Docker volume so your chats and settings can persist across container updates.

The following steps explain the entire process for a very new Windows user.

Step 1: Check Your Windows PC Before Installation

First, make sure your Windows computer is ready for Docker Desktop. You need a reasonably modern Windows installation with virtualization support available in the system. Docker Desktop uses WSL 2 on Windows for this type of setup.

If you have never used Docker before, do not worry. You do not need to understand containers in detail to follow this guide. Docker will provide the environment in which Open WebUI runs.

It is also important to understand that installing Open WebUI does not automatically install an AI model. Open WebUI provides the interface, while a service such as Ollama can provide local models.

Step 2: Install WSL 2 on Windows

Open the Start menu and search for PowerShell. Right-click Windows PowerShell and choose Run as administrator.

Enter the following command:

wsl –install

Press Enter and allow Windows to complete the installation. You may need to restart your computer.

WSL stands for Windows Subsystem for Linux. It allows Windows to run a Linux environment that can work with Docker Desktop.

Open WebUI’s documentation specifically recommends using WSL 2 with Docker Desktop for this Windows setup.

Step 3: Download and Install Docker Desktop

Now install Docker Desktop for Windows. During installation, use the WSL 2 backend when that option is presented.

After installation, launch Docker Desktop from the Start menu. Wait until Docker finishes starting.

You do not normally need to manually start a Linux server for Open WebUI. Docker Desktop manages the container environment for you.

The official Open WebUI documentation describes Docker as the officially supported and recommended installation method for most users.

Step 4: Enable WSL Integration in Docker Desktop

Open Docker Desktop and select Settings.

Go to Resources > WSL Integration. Make sure integration with your default WSL distribution is enabled.

If you have installed another WSL distribution and want to use it, you can enable integration for that distribution as well.

Click Apply or the equivalent option to save the settings.

This step allows Docker Desktop to work with your WSL environment.

Step 5: Open the WSL Terminal

Open the Start menu, type Ubuntu, and launch it if Ubuntu was installed with WSL.

You should see a terminal window.

This is where you can enter the Docker command needed to install and start Open WebUI.

Open WebUI’s Windows WSL instructions recommend running the Docker command from the WSL terminal.

Step 6: Download and Start Open WebUI

With Docker Desktop running, enter the following command in the WSL terminal:

docker run -d -p 3000:8080 –add-host=host.docker.internal:host-gateway -v open-webui:/app/backend/data -e WEBUI_SECRET_KEY=your-secret-key –name open-webui –restart always ghcr.io/open-webui/open-webui:main

Press Enter.

Docker will download the Open WebUI image and create a container called open-webui. The first download can take some time because Docker needs to obtain the required files.

The -p 3000:8080 portion makes Open WebUI available through port 3000 on your Windows computer. The -v open-webui:/app/backend/data portion creates persistent storage for Open WebUI’s data. The official documentation recommends keeping this volume because it stores things such as chats, users, and settings.

Step 7: Open ‘Open WebUI’ in Your Browser

Once the Docker command has completed, open Chrome, Edge, Firefox, or another modern browser.

Enter the following address in the address bar:

http://localhost:3000

Press Enter.

If everything is working correctly, the Open WebUI setup page should appear.

If the page does not immediately load, give Docker a little more time. You can also check the container logs with:

docker logs -f open-webui

The official documentation notes that you can wait for the application startup message when troubleshooting startup problems.

Step 8: Create Your Open WebUI Administrator Account

When Open WebUI starts for the first time, you should see an option to create an account.

Enter your desired name, email or username where requested, and password, depending on the version of Open WebUI you are using.

The first account created becomes the administrator account. The administrator can manage users and instance-wide settings.

Keep your administrator password somewhere safe because losing access to the administrator account can make managing the Open WebUI instance more difficult.

Step 9: Connect an AI Model Provider

At this point, Open WebUI is installed, but you need an AI model provider before you can have a useful conversation.

For local AI, one popular option is Ollama. If Ollama is running on the same Windows computer, Open WebUI can connect to it.

Open your Open WebUI settings and go to:

Settings > Admin > Connections

Look for Manage Ollama API Connections.

When Open WebUI is running inside Docker and Ollama is running on the Windows host, the connection can use:

http://host.docker.internal:11434

Open WebUI’s official documentation identifies this address for Docker users connecting to Ollama on the host machine.

Step 10: Download or Select an AI Model

After connecting to Ollama, you need at least one compatible model.

Open WebUI can provide model management through the Ollama connection. Depending on your configuration, you can enter a model name in the model selector and confirm the download.

Keep in mind that AI models can require significant disk space, RAM, and processing power. Larger models generally require more resources, so beginners should start with a model that their computer can reasonably handle.

Step 11: Start Your First AI Chat

Return to the main Open WebUI screen and click New Chat.

Use the model selector to choose one of your available models.

Type a simple question into the message box and press Enter.

Open WebUI will send the request to the selected model and display the response in the conversation.

You can continue asking questions in the same conversation, create additional chats, and switch between supported models when available.

Step 12: Upload a File and Ask Questions

Once your model and Open WebUI installation are working, you can experiment with document-based conversations.

Start a new chat and use the available file upload option to add a supported document.

You can then ask the AI questions about that document.

For example, you could upload a text document and ask the model to summarize it, explain a section, or find specific information. The exact capabilities can depend on the selected model and Open WebUI configuration.

Step 13: Install Open WebUI as a Desktop App

If you prefer using an application-like experience instead of opening a browser manually, Open WebUI can also be used as a Progressive Web App through supported browsers. Chrome and Edge can install supported web applications so that Open WebUI can appear like a desktop application.

Open WebUI also provides a separate desktop app option. However, for a beginner who wants the most straightforward Docker-based setup, opening http://localhost:3000 in a browser is usually sufficient.

FAQs

Is Open WebUI an AI model?

No. Open WebUI is an AI interface and platform, not an AI model itself. You connect it to services such as Ollama or compatible cloud and local AI providers.

Can I use Open WebUI without Ollama?

Yes. Open WebUI supports multiple providers, including OpenAI, Anthropic, OpenAI-compatible APIs, vLLM, and other backends. Ollama is only one option.

Where do I access Open WebUI after installing it?

With the Docker command used in this guide, open http://localhost:3000 in your web browser.

Does Open WebUI require Docker?

No. Docker is recommended for many users, but Open WebUI also provides Python, uv, Kubernetes, WSL, and desktop app installation options.

Does Open WebUI automatically include AI models?

No. Open WebUI itself does not provide the AI model you chat with. You need to connect a model provider such as Ollama or another supported provider.

Can I use Open WebUI offline?

Open WebUI is designed to support local and offline use when configured with suitable local AI models and services. Whether a particular feature works completely offline depends on the provider and feature being used.

What should I do if port 3000 is already being used?

If another application is already using port 3000, you can change the host-side port in the Docker command. For example, changing -p 3000:8080 to -p 3001:8080 would make Open WebUI available through port 3001 instead.

Summary

Installing Open WebUI on a Windows PC may initially sound complicated because the setup involves Docker, WSL, containers, and an AI model provider. However, once the basic components are installed, the process becomes much easier.

For a beginner, the general process is to install WSL 2, install Docker Desktop, enable WSL integration, run the official Open WebUI Docker command, and open http://localhost:3000. After creating your administrator account, connect Ollama or another supported AI provider, select an available model, and start chatting.

The most important thing to remember is that Open WebUI is the interface, while services such as Ollama or cloud APIs provide the actual AI models.

Once you understand this distinction, you can gradually explore more advanced features such as document conversations, knowledge bases, web search, plugins, tool calling, and other AI capabilities.