How to install Ollama and Open WebUI using docker compose
Install Ollama and Open WebUI
Stop and remove all containers which are already running.
docker stop $(docker ps -q); docker rm $(docker ps -q -a)Ollama Docker to run with NVIDIA GPU (–gpus=all)
docker run -d --gpus=all -v ollama:/root/.ollama -p 11434:11434 --name ollama ollama/ollamaYou might see these errors.
docker run -d --gpus=all -v ollama:/root/.ollama -p 11434:11434 --name ollama ollama/ollama
Unable to find image 'ollama/ollama:latest' locally
latest: Pulling from ollama/ollama
857cc8cb19c0: Pull complete
2c3df3d580d0: Pull complete
6014224340ef: Pull complete
fc86a784b436: Pull complete
Digest: sha256:bf65f12eb64051f9429d6348e65c2d4c5415981de10eefe3faa45f1078abb729
Status: Downloaded newer image for ollama/ollama:latest
fba9c1ad05b2deeace40580f5b347244930cae5c2d882a92704bab84ca040494
docker: Error response from daemon: driver failed programming external connectivity on endpoint ollama (747deb6a8acb6ef69f7e287c3719f379c94042da90f25a2397b3cc4286113e70): Error starting userland proxy: listen tcp4 0.0.0.0:11434: bind: address already in use.If see above errors, remove all containers as below and run again.
docker stop $(docker ps -q); docker rm $(docker ps -q -a)Open WebUI Docker
Run WebUI docker.
docker run -d -p 3000:8080 --env WEBUI_AUTH=False --add-host=host.docker.internal:host-gateway -v open-webui:/app/backend/data --name open-webui --restart always ghcr.io/open-webui/open-webui:mainOpen this URL, add a model such as “phi3”
http://localhost:3000/
Chooose the models from this URL depending on your PC’s RAM size.
https://www.ollama.com/library
Docker Compose
Configure Docker Compose to define ollama and open-webui in a single file (docker-compose.yml) and run with a single command (application) instead of running containers separately as before explanation.
Create docker-compose.yml file anywhere you like suck as “~/ollama”. This docker compose will create a separate storage (volume).
services:
ollama:
volumes:
- ollama:/root/.ollama
container_name: ollama
pull_policy: always
tty: true
restart: unless-stopped
image: ollama/ollama:${OLLAMA_DOCKER_TAG-latest}
environment:
- 'OLLAMA_HOST=0.0.0.0:11434'
# GPU support
deploy:
resources:
reservations:
devices:
- driver: ${OLLAMA_GPU_DRIVER-nvidia}
count: ${OLLAMA_GPU_COUNT-1}
capabilities:
- gpu
open-webui:
build:
context: .
args:
OLLAMA_BASE_URL: '/ollama'
dockerfile: Dockerfile
image: ghcr.io/open-webui/open-webui:${WEBUI_DOCKER_TAG-main}
container_name: open-webui
volumes:
- open-webui:/app/backend/data
depends_on:
- ollama
ports:
- ${OPEN_WEBUI_PORT-3000}:8080
environment:
- 'OLLAMA_BASE_URL=http://ollama:11434'
- 'WEBUI_AUTH=False'
extra_hosts:
- host.docker.internal:host-gateway
restart: unless-stopped
volumes:
ollama: {}
open-webui: {}These lines to use NVIDIA GPU
# GPU support
deploy:
resources:
reservations:
devices:
- driver: ${OLLAMA_GPU_DRIVER-nvidia}
count: ${OLLAMA_GPU_COUNT-1}
capabilities:
- gpuThis line to skip user registration screen.
- 'WEBUI_AUTH=False'To pull images and run containers
docker compose up -d
[+] Running 1/1
✔ ollama Pulled 2.7s
[+] Running 3/3
✔ Network test_default Created 0.1s
✔ Container ollama Started 0.5s
✔ Container open-webui Started 0.8sTo down containers
docker compose down
[+] Running 3/3
✔ Container open-webui Removed 1.7s
✔ Container ollama Removed 0.6s
✔ Network test_default Removed If you are running on WSL2, Ollama and Open WebUI will start again after your PC boot and WSL2 ubuntu terminal is started.
