5  User Interface (WebUI)

5.1 Create a Large Language Model Cluster

5.1.1 Create the Web UI Container

  • Create a volume called webui-data
docker volume create webui-data
  • Create the docker container with the Webui web graphical user interface by running the following command.
docker run -d -p 3000:8080 --add-host=host.docker.internal:host-gateway -v webui-data:/app/backend/data --network workshop_network --name webui --hostname webui --restart always ghcr.io/open-webui/open-webui:main

5.1.2 Initial Setup

Figure 5.1: WebUI Console
  • Create the Admin user.

WebUI Admin User creation
  • Accept the Release Notes.
Figure 5.2: WebUI Admin User creation
  • View the existing models already downloaded in the LLM engine. In this exercise, you should have llama3.2 already.
Figure 5.3: WebUI Admin User creation

5.1.3 Ask Questions

  • Pick your model and you are able to ask your first questions.
Figure 5.4: WebUI Admin User creation
  • View the answers.
Figure 5.5: WebUI Admin User creation

5.1.3.1 Connect WebUI with ChatGPT

  • You can also connect to ChatGPT API and use it’s models.

  • Navigate to the Admin Panel > Settings > Connections section.

  • Add your ChatGPT Key, similar to this one, in the last field of the form.

Important

This ChatGpt Key is no longer valid. It’s used here only as an example.

sk-proj-XlXRiWlfetudLBrRrP6B4C_RRjDpJu03LJ9smXeDz3p-gE7PLg-a1Td6qFCWoydnIL2pxhJL6dT3BlbkFJ5B2u
Figure 5.6: WebUI Admin User creation
  • Return to Workspace and you can now view that ChatGPT models are also listed.
Figure 5.7: WebUI Admin User creation