Skip to content

Configuring Ollama + Open WebUI on Linux

Installation of Open WebUI with Ollama Support

There are several methods to install Open WebUI with Ollama support. Here, we present three options for installation:

Option 1: Use the Bundled Ollama-Enabled Open WebUI

This is the simplest and recommended installation method. It uses a single container image that bundles Open WebUI with Ollama, allowing a streamlined setup with just one command. Choose the appropriate command based on your hardware environment:

With NVidia GPU and CUDA Support:

To leverage GPU resources, run the following command: 

sudo docker run -d -p 3000:8080 --gpus=all -v ollama:/root/.ollama -v open-webui:/app/backend/data --name open-webui --restart always ghcr.io/open-webui/open-webui:cuda

With GPU Support (Without CUDA):

To utilize GPU resources, run the following command: 

sudo docker run -d -p 3000:8080 --gpus=all -v ollama:/root/.ollama -v open-webui:/app/backend/data --name open-webui --restart always ghcr.io/open-webui/open-webui:ollama

Only CPU (No GPU):

If you do not have a GPU, use the following command: 

sudo docker run -d -p 3000:8080 -v ollama:/root/.ollama -v open-webui:/app/backend/data --name open-webui --restart always ghcr.io/open-webui/open-webui:ollama

Verify Docker Instance is Running:

To check if the Docker instance is running, use the following command: 

$ sudo docker ps

Expected output: 

CONTAINER ID   IMAGE                                  COMMAND           CREATED             STATUS             PORTS                                       NAMES
984c0d7006ef   ghcr.io/open-webui/open-webui:ollama   "bash start.sh"   About an hour ago   Up About an hour   0.0.0.0:3000->8080/tcp, :::3000->8080/tcp   open-webui

Option 2: Install Open WebUI with Separate Ollama Instance

This option allows you to install Ollama first and then run Open WebUI using a separate instance of Ollama.

Install Ollama:

Run the following command to install Ollama: 

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

Start Open WebUI:

Once Ollama is installed, use the following command to start Open WebUI: 

sudo 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

Connect to Ollama on a Remote Server:

If you want to connect to Ollama running on a remote server, change the OLLAMA_BASE_URL to the remote server's URL: 

sudo docker run -d -p 3000:8080 -e OLLAMA_BASE_URL=https://your_domain -v open-webui:/app/backend/data --name open-webui --restart always ghcr.io/open-webui/open-webui:main

Run Open WebUI with Nvidia GPU Support:

If you're using an Nvidia GPU, run the following command: 

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

Verify Docker Instance is Running:

To check if the Docker instance is running, use the following command: 

$ sudo docker ps

Option 3: Install Using Docker Compose

This option involves cloning the Open WebUI GitHub repository and using Docker Compose for installation.

Clone the Repository:

First, clone the Open WebUI GitHub repository: 

git clone https://github.com/open-webui/open-webui
cd open-webui

Select the Appropriate Docker Compose File:

Choose the appropriate Docker Compose file based on your hardware:

  • docker-compose.amdgpu.yaml — For AMD GPUs
  • docker-compose.api.yaml — API-only setup
  • docker-compose.data.yaml — For data services
  • docker-compose.gpu.yaml — For Nvidia GPUs
  • docker-compose.yaml — Default configuration

Start Docker Environment:

Run the following command to start the Docker Compose environment: 

sudo docker compose -f docker-compose.yaml up

Verify Docker Instance is Running:

To check if the Docker instance is running, use the following command: 

$ sudo docker ps

Expected output: 

CONTAINER ID   IMAGE                                  COMMAND           CREATED             STATUS             PORTS                                       NAMES
984c0d7006ef   ghcr.io/open-webui/open-webui:ollama   "bash start.sh"   About an hour ago   Up About an hour   0.0.0.0:3000->8080/tcp, :::3000->8080/tcp   open-webui

Accessing the User Interface

Once the setup is complete, open your web browser and navigate to:

http://<your_server_ip>:3000

Info

  • The first time you start, there might be a white screen for a few minutes. This is because the system is trying to make a request to the openai.com address. To fix this, simply disable the OpenAI API and enable the Ollama API in the settings.
  • If, after startup, the local Ollama model you pulled cannot be found, go to the admin panel and correctly input the local Ollama address.

ref: a CSDN blog by vvw&

Comments