OpenWebUI Deploying the Ollama Visualization Dialog Interface

I. Deploying with Node.js

1. Install Node.js

Download and install Node.js Tools: https://www.nodejs.com.cn/download.html

Set the mirror source, for example, using the following mirror source.

npm config set registry http://mirrors.cloud.tencent.com/npm/

2. Clone the ollama-webui repository

Clone the ollama-webui repository in the terminal and switch the directory of the ollama-webui code.

git clone https://github.com/ollama-webui/ollama-webui-lite.git
cd ollama-webui-lite

3. Installation of dependencies

Install the dependency in the terminal.

npm ci

4. Operation

npm run dev

strike (on the keyboard) http://localhost:3000/ , open the page.

Deploying with Docker

This subsection deploys open-webui using Docker on Windows.

Reference Tutorial Ollama Installation and Configuration for WindowsTo install Ollama in advance, use the following command in the terminal to pull the ghcr.io/open-webui/open-webui:main Mirror it and run it.

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

The following result appears in the command terminal, indicating a successful run.

After the command is run successfully, the following command will appear under Docker's Container open-webui, as shown in the figure below.

After that click on Port Access as shown below. You can also directly type in your browser http://localhost:3000/ Access.

When you open it, it will appear that you are logged in to Open WebUI, just register for an account.

After selecting the model, you can start the dialog.

The following is a guide to using the qwen2:1.5b An example of a dialog.

If you want to explore more features you can refer to: https://github.com/open-webui/open-webui

Reference warehouse address: https://github.com/ollama-webui/ollama-webui-lite