No GPU device found in OpenVINO. Falling back to CPU.
Hello everyone, I am becoming desperate because I am unable to enable hardware accelerated Machine Learning. I tried to edit my Docker compose file as shown (can only use one in Portainer), but it simply doesnt work for me. I am running a DXP4800 Plus NAS from UGreen (with Ugos OS) with the latest Immich-Release, which should have an Intel UHD integrated GPU and therefore "OpenVino", right?
Here is my Docker file, could you please take a look and tell me, what I am doing wrong?
https://nopaste.net/NCvarOgldQ
The logs look like this for immich-learning:
[05/09/25 11:33:05] INFO Starting gunicorn 23.0.0
[05/09/25 11:33:05] INFO Listening at: http://[::]:3003 (8)
[05/09/25 11:33:05] INFO Using worker: immich_ml.config.CustomUvicornWorker [05/09/25 11:33:05] INFO Booting worker with pid: 9
[05/09/25 11:33:07] INFO Started server process [9]
[05/09/25 11:33:07] INFO Waiting for application startup.
[05/09/25 11:33:07] INFO Created in-memory cache with unloading after 300s
of inactivity.
[05/09/25 11:33:07] INFO Initialized request thread pool with 6 threads.
[05/09/25 11:33:07] INFO Application startup complete.
[05/09/25 11:33:08] INFO Loading detection model 'buffalo_l' to memory
[05/09/25 11:33:08] WARNING No GPU device found in OpenVINO. Falling back to
CPU.
[05/09/25 11:33:08] INFO Setting execution providers to
['CPUExecutionProvider'], in descending order of
preference
Thank you very much in advance! I really love Immich! Tom
[05/09/25 11:33:05] INFO Listening at: http://[::]:3003 (8)
[05/09/25 11:33:05] INFO Using worker: immich_ml.config.CustomUvicornWorker [05/09/25 11:33:05] INFO Booting worker with pid: 9
[05/09/25 11:33:07] INFO Started server process [9]
[05/09/25 11:33:07] INFO Waiting for application startup.
[05/09/25 11:33:07] INFO Created in-memory cache with unloading after 300s
of inactivity.
[05/09/25 11:33:07] INFO Initialized request thread pool with 6 threads.
[05/09/25 11:33:07] INFO Application startup complete.
[05/09/25 11:33:08] INFO Loading detection model 'buffalo_l' to memory
[05/09/25 11:33:08] WARNING No GPU device found in OpenVINO. Falling back to
CPU.
[05/09/25 11:33:08] INFO Setting execution providers to
['CPUExecutionProvider'], in descending order of
preference
Thank you very much in advance! I really love Immich! Tom
27 Replies
:wave: Hey @TomKay,
Thanks for reaching out to us. Please carefully read this message and follow the recommended actions. This will help us be more effective in our support effort and leave more time for building Immich :immich:.
References
- Container Logs:
docker compose logs docs
- Container Status: docker ps -a docs
- Reverse Proxy: https://immich.app/docs/administration/reverse-proxy
- Code Formatting https://support.discord.com/hc/en-us/articles/210298617-Markdown-Text-101-Chat-Formatting-Bold-Italic-Underline#h_01GY0DAKGXDEHE263BCAYEGFJA
Checklist
I have...
1. :ballot_box_with_check: verified I'm on the latest release(note that mobile app releases may take some time).
2. :ballot_box_with_check: read applicable release notes.
3. :ballot_box_with_check: reviewed the FAQs for known issues.
4. :ballot_box_with_check: reviewed Github for known issues.
5. :ballot_box_with_check: tried accessing Immich via local ip (without a custom reverse proxy).
6. :ballot_box_with_check: uploaded the relevant information (see below).
7. :ballot_box_with_check: tried an incognito window, disabled extensions, cleared mobile app cache, logged out and back in, different browsers, etc. as applicable
(an item can be marked as "complete" by reacting with the appropriate number)
Information
In order to be able to effectively help you, we need you to provide clear information to show what the problem is. The exact details needed vary per case, but here is a list of things to consider:
- Your docker-compose.yml and .env files.
- Logs from all the containers and their status (see above).
- All the troubleshooting steps you've tried so far.
- Any recent changes you've made to Immich or your system.
- Details about your system (both software/OS and hardware).
- Details about your storage (filesystems, type of disks, output of commands like fdisk -l and df -h).
- The version of the Immich server, mobile app, and other relevant pieces.
- Any other information that you think might be relevant.
Please paste files and logs with proper code formatting, and especially avoid blurry screenshots.
Without the right information we can't work out what the problem is. Help us help you ;)
If this ticket can be closed you can use the /close command, and re-open it later if needed.
Successfully submitted, a tag has been added to inform contributors. :white_check_mark:Does your ugos is have a driver for that gpu?
Kernel 5 15 or higher is required
I am not sure, if something special must be installed or not. Other people with the smaller, 2 bay NAS got it running somehow. Where can I check it to provide further information?
Tom@BKA-Server-2:~$ uname -r
6.1.27
Does this help ?
Linux version 6.1.27 (ugreen@debian) (gcc (Debian 12.2.0-14) 12.2.0, GNU ld (GNU Binutils for Debian) 2.40) #36 SMP PREEMPT_DYNAMIC Mon Apr 14 23:38:49 CST 2025
what's in /dev/dri ?
drwxr-xr-x 2 root root 80 May 7 13:38 by-path
crw-rw---- 1 root video 226, 0 May 7 13:38 card0
crw-rw---- 1 root render 226, 128 May 7 13:38 renderD128
- /volume1/immich/matplotlib:/matplotlib:rw environment: - MPLCONFIGDIR=/matplotlibI see you're a Marius fan I think
no-new-privileges might be stopping your container here, not sureyup, it may be related to permissions. I would also remove the user directive
ah right, the users might not be in the render/video group
Could work to add:
Or their equivalent GID
What should I do with ".... etc ..." 😄 Sorry, I am kind of a noob for Docker actually.
This
group_add:
- video
- render
I add under immich-learning?
I just wanted to show the indentation level
Yes
type
cat /etc/group | grep -e render -e video
replace render/video in the yaml with the number between "
so -render for instance will be - "44" or suchHOLY MOLY 😄
I think it worked
it seems to utilize 25% of GPU now 🙂
Face detection count drops fast 😄 Niceeee ... thank you so much guys! I dont quite understand why this had to be done, but its awesome to see it working.
The user the container runs under did not have priveliges to access the GPU
if you add the GPU group to the container then it does
Would there be a way to utilize even more than 25% ? it now seems to be the maximum currently
Keep in mind that there is a limit to what can run on the GPU
it still needs to queue the job, fetch the file ...
You can increase concurrency for ML tasks in the admin settings, but don't go nuts
if the default is 1 or 2, then 4 or maybe 8 might be the limit to what is productive
its running quite fast already 😄 did like 4000 faces since i posted this screenshot
any more and you just bottleneck it with other htings
Check the logs too to make sure it's not erroring, those also run fast 🙃
no errors on immich-learning...
hurray
face detection goes brrrrrrrrrrrr 😄
Awesome, will get a supporter package the next days. This community rocks ❤️
Nice 🙂