How to Install Locally and Securely DeepSeek-R1 with Graphics User Interface (Open WebUI) on Linux Ubuntu

In our previous tutorial on DeepSeek-R1, we explained how to install and run DeepSeek-R1 locally by using Ollama and a terminal window. Running DeepSeek and similar Large Language Models (LLMs) by using Ollama and a terminal is a first step toward building an AI application. Namely, you would first like to test the model and see if it can work on your hardware, and then, in the next step, you would embed the AI model in your application. For that purpose, you would most likely use Python or a similar language which enables rapid prototyping.

However, some users would just like to run DeepSeek-R1 locally with a graphics user interface (GUI). There are several approaches for running distilled versions of DeepSeek in a GUI. The first approach that we covered in the previous tutorial is to write your own GUI by using Streamlit. This approach is for more advanced users who want to build their own AI application. However, there is a simple approach which is more suitable for users only interested in directly using the model and not in embedding the model in some other application. For that purpose, you can use WebUI to securely and locally run models. In this tutorial, we explain a step by step procedure that will enable you to install and run distilled versions of DeepSeek by using Open WebUI on Linux Ubuntu. Open WebUI is a relatively simple framework and GUI for running models locally. The YouTube tutorial is given below.

The GUI interface of the WebUI with the DeepSeek-R1 7B model is shown in the figure below.

Installation Procedure

The first step is to install Ollama. For that purpose, open the Linux Ubuntu terminal, and then update and upgrade all the packages, and install curl:

sudo apt update && sudo apt upgrade 
sudo apt install curl 
curl --version

Next, enable connections to

For that purpose, go to the official Ollama website

https://www.ollama.com

and then click on download, and then select your operating system. As the result the curl installation command will be generated (see the figures below).

The generated installation command is

curl -fsSL https://ollama.com/install.sh | sh

Execute this command in the terminal window and Ollama should be installed.

The next step is to verify the Ollama installation. To do that, in your web browser type the following address

http://localhost:11434/

If Ollama is properly installed, the generated message should look like this

Another verification step is to open a terminal and to type

ollama list

If Ollama is properly installed, then the response will look like this

The next step is to install the model. In our tutorial, we will use a distilled version of DeepSeek-R1. To install the model, go to the Ollama website and search for deepseek-r1

Click on the model, and then click on 7B model in the drop down menu as shown in the figure below.

as the result the installation command will be generated (right to the drop down menu). In our case, the installation command is

ollama run deepseek-r1:7b

This command will download, install and run the model. However, for the time being, we just want to download and install the model. To do that, change the command to

ollama pull deepseek-r1:7b

and execute the command in the terminal. This command will download the model. After the model is downloaded, open the terminal and type

ollama list

This will list all the models. In our case, the output should look like this

NAME              ID              SIZE      MODIFIED    
deepseek-r1:7b    0a8c26691023    4.7 GB    3 hours ago 

To test the model, run it like this

ollama run deepseek-r1:7b

You can test the model in order to confirm it is working. To exit Ollama, press and hold CTRL+d

The next step is to create a workspace folder, create a Python virtual environment, install the libraries, and run Open-WebUI. To create a workspace folder, type

cd ~
mkdir testWebUI
cd testWebUI

Next, check if Python is installed on your system and check your Python version by typing

which python3
python3 --version 

In our case, we are using Python 3.12, and we suggest you to use the same version.

To create and activate the Python virtual environment, type this

sudo apt install python3.12-venv
python3 -m venv env1
source env1/bin/activate

Then, you just need to install a single library called open-webui. To do that, type in the created Python virtual environment

pip install open-webui

Finally, to run Open-WebUI, type this

open-webui serve

After Open-WebUI is started, go to the web browser and enter this address

http://localhost:8080

This will start Open-WebUI. After the GUI is started enter your information and you can start the model (for more details see the YouTube tutorial). The GUI should look like this: