CentOS Stream 9
Sponsored Link

Open WebUI : Install2024/06/04

 

Install Open WebUI which allows you to run LLM on Web UI.

Open WebUI can be easily installed with pip3, however in this example, we will start it in a container.

[1]

Install Podman, refer to here.

[2]

Install Ollama, refer to here.

[3] Pull and start the Open WebUI container image.
[root@dlp ~]#
podman pull ghcr.io/open-webui/open-webui:main

[root@dlp ~]#
podman images

REPOSITORY                     TAG         IMAGE ID      CREATED      SIZE
ghcr.io/open-webui/open-webui  main        a48d0a909833  3 hours ago  3.58 GB

[root@dlp ~]#
podman run -d -p 3000:8080 --add-host=host.containers.internal:host-gateway -v open-webui:/app/backend/data --name open-webui --restart always ghcr.io/open-webui/open-webui:main
[root@dlp ~]#
podman ps

CONTAINER ID  IMAGE                               COMMAND        CREATED        STATUS        PORTS                   NAMES
efc2886a7f58  ghcr.io/open-webui/open-webui:main  bash start.sh  6 seconds ago  Up 6 seconds  0.0.0.0:3000->8080/tcp  open-webui
[4] Start any web browser on client computer and access to Open WebUI.
When you access the application, the following screen will appear. For initial accessing to the application, click [Sign up] to register as a user.
[5] Enter the required information and click [Create Account]. The first user to register will automatically become an administrator account.
[6] Once your account is created, the Open WebUI default page will be displayed.
[7] Next time, you can log in with your registered email address and password.
[8] For second and subsequent users, when you register by clicking [Sign up], the account will be in Pending status and must be approved by an administrator account.
[9] To approve a new user, log in with your administrator account and click Admin Panel from the menu in the top right.
[10] If you click [pending] on a user in [Pending] status, that user will be approved. If you click [user] again, that user can be promoted to [Admin].
[11] To use Chat, select the model you have loaded into Ollama from the menu at the top, enter a message in the box below, and you will receive a reply.
Matched Content