Immich – Stable GPU Deployment Guide (v1.105.0)

Stable Pinned Versions Docker NVIDIA GPU TensorRT 2025

1. Install Docker

sudo apt update
sudo apt install -y ca-certificates curl gnupg

sudo install -m 0755 -d /etc/apt/keyrings
curl -fsSL https://download.docker.com/linux/ubuntu/gpg | \
  sudo gpg --dearmor -o /etc/apt/keyrings/docker.gpg

echo \
  "deb [arch=$(dpkg --print-architecture) \
  signed-by=/etc/apt/keyrings/docker.gpg] \
  https://download.docker.com/linux/ubuntu \
  $(lsb_release -cs) stable" | \
  sudo tee /etc/apt/sources.list.d/docker.list > /dev/null

sudo apt update
sudo apt install -y docker-ce docker-ce-cli containerd.io docker-buildx-plugin docker-compose-plugin

2. Install NVIDIA GPU Drivers

sudo apt install -y nvidia-driver-535
sudo reboot

3. Install NVIDIA Container Toolkit

curl -fsSL https://nvidia.github.io/libnvidia-container/gpgkey | \
  sudo gpg --dearmor -o /usr/share/keyrings/nvidia-container-toolkit.gpg

curl -s -L https://nvidia.github.io/libnvidia-container/stable/deb/nvidia-container-toolkit.list | \
  sed 's#deb https://#deb [signed-by=/usr/share/keyrings/nvidia-container-toolkit.gpg] https://#' | \
  sudo tee /etc/apt/sources.list.d/nvidia-container-toolkit.list

sudo apt update
sudo apt install -y nvidia-container-toolkit

sudo nvidia-ctk runtime configure --runtime=docker
sudo systemctl restart docker

4. Verify GPU passthrough

docker run --rm --gpus all nvidia/cuda:12.2.0-base nvidia-smi

5. Create Immich directory

mkdir -p /opt/immich
cd /opt/immich

6. Create .env

IMMICH_VERSION=v1.105.0
DB_USERNAME=immich
DB_PASSWORD=immich
DB_DATABASE_NAME=immich

7. Stable docker-compose.yml (Pinned Versions)

services:
  immich-server:
    container_name: immich_server
    image: ghcr.io/immich-app/immich-server:v1.105.0
    depends_on:
      - redis
      - database
    env_file:
      - .env
    volumes:
      - immich_upload:/usr/src/app/upload
    restart: always

  immich-machine-learning:
    container_name: immich_machine_learning
    image: ghcr.io/immich-app/immich-machine-learning:v1.105.0-cuda
    env_file:
      - .env
    volumes:
      - model-cache:/cache
    deploy:
      resources:
        reservations:
          devices:
            - driver: nvidia
              count: 1
              capabilities: [gpu]
    restart: always

  redis:
    container_name: immich_redis
    image: redis:7.2.4
    restart: always

  database:
    container_name: immich_postgres
    image: tensorchord/pgvecto-rs:pg14-v0.2.0
    environment:
      POSTGRES_USER: ${DB_USERNAME}
      POSTGRES_PASSWORD: ${DB_PASSWORD}
      POSTGRES_DB: ${DB_DATABASE_NAME}
    volumes:
      - pgdata:/var/lib/postgresql/data
    restart: always

volumes:
  immich_upload:
  pgdata:
  model-cache:

8. Start Immich

docker compose up -d

9. Access Immich

• Local / Signal Raider: http://localhost:2283
• Johannes: via HAProxy backend

10. GPU verification inside Immich

• Settings → Machine Learning
• CUDA should show as available
• Indexing photos should use the GPU heavily

Versioning Philosophy

Immich evolves quickly, and floating tags can introduce unexpected changes:

• schema migrations may change
• Postgres vector extensions may update
• CUDA builds may shift
• ML models may be replaced
• container tags may be removed or made private

By pinning versions:

• deployments become reproducible
• tutorials remain accurate over time
• users avoid surprise breakage
• debugging becomes consistent
• future upgrades can be planned instead of forced

Once validated, this section will be updated to:

• ✔ Fully tested on Signal Raider
• ✔ Production‑ready for Johannes

This approach ensures that every guide on i9.nu remains trustworthy, predictable, and aligned with real‑world deployments.