Tried to download models after connecting to Open Web UI directly with IP and Port (to eliminate any possible issue with reverse proxy)
Added DNS entries to docker-compose.yml to solve the issue, but didn't work
Tried purging models on Ubuntu Server with rm -rf /volume1/docker/ollama/models/* and then redownloading models. Didn't Fix. Some models download (like deepseek-r1:1.5b and deepseek-r1:8b) but larger models continue to fail (like deepseek-r1:14b and deepseek-r1:32b).
Tried manually adding nameserver entry to Ollama container by adding echo "nameserver 8.8.8.8" > /etc/resolv.conf via Exec console. Didn't fix.
Logs & Screenshots
Additional Information
No response
Originally created by @jlar0che on GitHub (Mar 6, 2025).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/11306
### Check Existing Issues
- [x] I have searched the existing issues and discussions.
### Installation Method
Docker
### Open WebUI Version
v0.5.20
### Ollama Version (if applicable)
v0.5.13
### Operating System
Ubuntu Server 22.04.2
### Browser (if applicable)
Firefox v136.0, Chrome v133.0.6943.142
### Confirmation
- [x] I have read and followed all instructions in `README.md`.
- [x] I am using the latest version of **both** Open WebUI and Ollama.
- [x] I have checked the browser console logs.
- [x] I have checked the Docker container logs.
- [x] I have listed steps to reproduce the bug in detail.
### Expected Behavior
- [x] Click the "down arrow" next to the Models in the web ui
- [x] Type in Model name (e.g. deepseek-r1:32b)
- [x] Click on the `Pull "_model-name_" from Ollama.com` button
- [ ] Wait for the download to successfully complete 100% and the model gets added to the available models list
### Actual Behavior
Hash verification process at the end of the download fails (see attached photo). Model never completes download.
``
### Steps to Reproduce
1. Setup Ubuntu Server 22.04.2
2. Verify enough disk space with `df -h` (see screenshot below)
3. Install Docker via apt repository for Ubuntu as per instructions here: https://docs.docker.com/engine/install/ubuntu/
4. Install Open Web UI with Ollama support (see docker-compose.yml file below)
```
services:
ollama:
volumes:
- /volume1/docker/ollama:/root/.ollama:rw
container_name: ollama
pull_policy: always
tty: true
restart: always
image: ollama/ollama:latest
environment:
- NVIDIA_VISIBLE_DEVICES=all
- NVIDIA_DRIVER_CAPABILITIES=compute,utility
dns:
- 8.8.8.8
- 1.1.1.1
deploy:
resources:
reservations:
devices:
- driver: nvidia
count: all
capabilities: [gpu]
open-webui:
image: 'ghcr.io/open-webui/open-webui:cuda' #or use ":main"
container_name: open-webui
restart: always
volumes:
- '/volume1/docker/open-webui:/app/backend/data:rw'
depends_on:
- ollama
environment:
- 'OLLAMA_BASE_URL=http://ollama:11434'
- 'WEBUI_SECRET_KEY='
ports:
- '3808:8080'
extra_hosts:
- host.docker.internal:host-gateway
dns:
- 8.8.8.8
- 1.1.1.1
volumes:
ollama: {}
open-webui: {}
```
5. Attempt to download Models
6. Tried to download models after connecting to Open Web UI directly with IP and Port (to eliminate any possible issue with reverse proxy)
7. Added DNS entries to docker-compose.yml to solve the issue, but didn't work
8. Tried purging models on Ubuntu Server with `rm -rf /volume1/docker/ollama/models/*` and then redownloading models. Didn't Fix. Some models download (like deepseek-r1:1.5b and deepseek-r1:8b) but larger models continue to fail (like deepseek-r1:14b and deepseek-r1:32b).
9. Tried manually adding nameserver entry to Ollama container by adding `echo "nameserver 8.8.8.8" > /etc/resolv.conf` via Exec console. Didn't fix.
### Logs & Screenshots
### Additional Information
_No response_
GiteaMirror
added the bug label 2026-05-18 00:34:11 -05:00
As i see that you are on console, you can try for download the model directly from ollama on terminal: ollama pull deepseek-r1:8b
(and you can see what happend when it is downloading and its duration, if there are retries,..., you can look for other models as well in https://ollama.com/search )
<!-- gh-comment-id:2704550983 -->
@rgaricano commented on GitHub (Mar 6, 2025):
As i see that you are on console, you can try for download the model directly from ollama on terminal: `ollama pull deepseek-r1:8b`
(and you can see what happend when it is downloading and its duration, if there are retries,..., you can look for other models as well in https://ollama.com/search )
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
Originally created by @jlar0che on GitHub (Mar 6, 2025).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/11306
Check Existing Issues
Installation Method
Docker
Open WebUI Version
v0.5.20
Ollama Version (if applicable)
v0.5.13
Operating System
Ubuntu Server 22.04.2
Browser (if applicable)
Firefox v136.0, Chrome v133.0.6943.142
Confirmation
README.md.Expected Behavior
Pull "_model-name_" from Ollama.combuttonActual Behavior
Hash verification process at the end of the download fails (see attached photo). Model never completes download.
Steps to Reproduce
df -h(see screenshot below)rm -rf /volume1/docker/ollama/models/*and then redownloading models. Didn't Fix. Some models download (like deepseek-r1:1.5b and deepseek-r1:8b) but larger models continue to fail (like deepseek-r1:14b and deepseek-r1:32b).echo "nameserver 8.8.8.8" > /etc/resolv.confvia Exec console. Didn't fix.Logs & Screenshots
Additional Information
No response
@rgaricano commented on GitHub (Mar 6, 2025):
As i see that you are on console, you can try for download the model directly from ollama on terminal:
ollama pull deepseek-r1:8b(and you can see what happend when it is downloading and its duration, if there are retries,..., you can look for other models as well in https://ollama.com/search )