ML on Intel n100

[serj-47]

I have searched the existing issues, both open and closed, to make sure this is not a duplicate report.

• Yes

The bug

not work hw.accel ML

The OS that Immich Server is running on

Debian 12

Version of Immich Server

2.4.1

Version of Immich Mobile App

2.4.1

Platform with the issue

• Server
• Web
• Mobile

Device make and model

No response

Your docker-compose.yml content

#
# WARNING: To install Immich, follow our guide: https://docs.immich.app/install/docker-compose
#
# Make sure to use the docker-compose.yml of the current release:
#
# https://github.com/immich-app/immich/releases/latest/download/docker-compose.yml
#
# The compose file on main may not be compatible with the latest release.

name: immich

services:
  immich-server:
    container_name: immich_server
    image: ghcr.io/immich-app/immich-server:${IMMICH_VERSION:-release}
    # extends:
    #   file: hwaccel.transcoding.yml
    #   service: cpu # set to one of [nvenc, quicksync, rkmpp, vaapi, vaapi-wsl] for accelerated transcoding
    volumes:
      # Do not edit the next line. If you want to change the media storage location on your system, edit the value of UPLOAD_LOCATION in the .env file
      - ${UPLOAD_LOCATION}:/data
      - /etc/localtime:/etc/localtime:ro
    env_file:
      - .env
    ports:
      - '2283:2283'
    depends_on:
      - redis
      - database
    restart: always
    healthcheck:
      disable: false

  immich-machine-learning:
    container_name: immich_machine_learning
    # For hardware acceleration, add one of -[armnn, cuda, rocm, openvino, rknn] to the image tag.
    # Example tag: ${IMMICH_VERSION:-release}-cuda
    image: ghcr.io/immich-app/immich-machine-learning:${IMMICH_VERSION:-release}-openvino
    extends: # uncomment this section for hardware acceleration - see https://docs.immich.app/features/ml-hardware-acceleration
     file: hwaccel.ml.yml
     service: openvino # set to one of [armnn, cuda, rocm, openvino, openvino-wsl, rknn] for accelerated inference - use the `-wsl` version for WSL2 where applicable
    volumes:
      - model-cache:/cache
    env_file:
      - .env
    restart: always
    healthcheck:
      disable: false

  redis:
    container_name: immich_redis
    image: docker.io/valkey/valkey:8@sha256:81db6d39e1bba3b3ff32bd3a1b19a6d69690f94a3954ec131277b9a26b95b3aa
    healthcheck:
      test: redis-cli ping || exit 1
    restart: always

  database:
    container_name: immich_postgres
    image: ghcr.io/immich-app/postgres:14-vectorchord0.4.3-pgvectors0.2.0@sha256:bcf63357191b76a916ae5eb93464d65c07511da41e3bf7a8416db519b40b1c23
    environment:
      POSTGRES_PASSWORD: ${DB_PASSWORD}
      POSTGRES_USER: ${DB_USERNAME}
      POSTGRES_DB: ${DB_DATABASE_NAME}
      POSTGRES_INITDB_ARGS: '--data-checksums'
      # Uncomment the DB_STORAGE_TYPE: 'HDD' var if your database isn't stored on SSDs
      # DB_STORAGE_TYPE: 'HDD'
    volumes:
      # Do not edit the next line. If you want to change the database storage location on your system, edit the value of DB_DATA_LOCATION in the .env file
      - ${DB_DATA_LOCATION}:/var/lib/postgresql/data
    shm_size: 128mb
    restart: always

volumes:
  model-cache:

Your .env content

default

Reproduction steps

I deployed the containers on my server and started the task of defining the text on the photo. I run intel_gpu_top and it shows the gpu is not being used in any way.

hwaccel.ml.yml - default

docker logs immich_machine_learning
[01/07/26 18:20:51] INFO     Starting gunicorn 23.0.0
[01/07/26 18:20:51] INFO     Listening at: http://[::]:3003 (11)
[01/07/26 18:20:51] INFO     Using worker: immich_ml.config.CustomUvicornWorker
[01/07/26 18:20:51] INFO     Booting worker with pid: 15
[01/07/26 18:20:53] INFO     Started server process [15]
[01/07/26 18:20:53] INFO     Waiting for application startup.
[01/07/26 18:20:53] INFO     Created in-memory cache with unloading after 300s
                             of inactivity.
[01/07/26 18:20:53] INFO     Initialized request thread pool with 4 threads.
[01/07/26 18:20:53] INFO     Application startup complete.
[01/07/26 18:32:21] INFO     Loading detection model 'PP-OCRv5_mobile' to memory
[01/07/26 18:32:21] INFO     Setting execution providers to
                             ['OpenVINOExecutionProvider',
                             'CPUExecutionProvider'], in descending order of
                             preference
[01/07/26 18:32:21] INFO     Loading recognition model 'PP-OCRv5_mobile' to
                             memory
[01/07/26 18:32:21] INFO     Setting execution providers to
                             ['OpenVINOExecutionProvider',
                             'CPUExecutionProvider'], in descending order of
                             preference
[INFO] 2026-01-07 18:32:22,069 [RapidOCR] base.py:22: Using engine_name: onnxruntime

docker exec -t immich_machine_learning ls -la /dev/dri
total 0
drwxr-xr-x 2 root root        80 Jan  7 18:20 .
drwxr-xr-x 7 root root       380 Jan  7 18:20 ..
crw-rw---- 1 root video 226,   0 Jan  7 18:20 card0
crw-rw---- 1 root   105 226, 128 Jan  7 18:20 renderD128

Relevant log output

Additional information

No response

[serj-47]

Guys, well, has no one raised hw accel machine learning on intel n100? I have updated all packages on debian, the kernel to 6.1.0-42. The video driver is installed from apt, there may be a problem with it, it's not very old, though, or maybe for hw machine learning to work, I also need to enable hardware transcoding, but I'm not sure.

apt show intel-media-va-driver
Package: intel-media-va-driver
Version: 23.1.1+dfsg1-1
Priority: optional
Section: video
Source: intel-media-driver
Maintainer: Debian Multimedia Maintainers <debian-multimedia@lists.debian.org>
Installed-Size: 14,5 MB
Provides: va-driver
Depends: libva-driver-abi-1.17, libc6 (>= 2.34), libgcc-s1 (>= 3.0), libigdgmm12 (>= 22.3.3+ds1), libstdc++6 (>= 11)

[serj-47]

The problem was that I deployed the containers first and then edited docker-compose, and for some reason all time the machine_learning container was used by default instead of the machine_learning-openvino container.