imagegenius/obico

Obico is a community-built, open-source smart 3D printing platform used by makers, enthusiasts, and tinkerers around the world.

Variants

Obico Server does not publish semver releases. This image pins the upstream release branch commit in docker-bake.hcl.

NVIDIA GPU Acceleration

Use the cuda tag with the NVIDIA Container Toolkit and pass through the GPU.

services:
  obico:
    image: ghcr.io/imagegenius/obico:cuda
    deploy:
      resources:
        reservations:
          devices:
            - driver: nvidia
              capabilities: [gpu]

Requirements

• Redis: External or via docker mod (see below).
• HOST_IP: Set this to the host, IP:port, or DNS name used to access Obico.

Docker Mod for Redis

• Set DOCKER_MODS=imagegenius/mods:universal-redis
• Set REDIS_URL=redis://localhost:6379

Usage

Docker Compose

---
services:
  obico:
    image: ghcr.io/imagegenius/obico:latest
    container_name: obico
    environment:
      - PUID=1000
      - PGID=1000
      - TZ=Etc/UTC
      - REDIS_URL=redis://192.168.1.x:6379
      - HOST_IP=192.168.1.x:3334
    volumes:
      - path_to_appdata:/config
    ports:
      - 3334:3334
    restart: unless-stopped

  redis:
    image: redis:7-alpine
    container_name: redis
    ports:
      - 6379:6379

Parameters

Application Setup

After first start, configure the Django site domain so static assets and links resolve correctly. Follow Obico's server configuration guide, especially "Login as Django admin" and "Configure Django site".

User / Group IDs & umask

Set PUID=1000 PGID=1000 to match volume ownership on the host (id user to find yours). Optionally UMASK=022 (works subtractively, not additively).

Updating

docker pull ghcr.io/imagegenius/obico:latest
docker stop obico && docker rm obico
# recreate with the same docker run parameters
docker image prune  # optional: remove dangling images

Or with compose: docker compose pull && docker compose up -d.

Support

• Issues: https://github.com/imagegenius/docker-obico/issues
• Obico: https://www.obico.io/

How this image is built

This repo is built with GitHub Actions, based on the workflow shape from home-operations/containers.

• The container starts from linuxserver/docker-baseimage-ubuntu.
• Obico Server is fetched from the upstream release branch commit pinned in docker-bake.hcl.
• Darknet is built from the same upstream-pinned AlexeyAB commit used by Obico's current ML base image.
• The backend, ML API, frontend, Moonraker, model weights, Darknet libraries, and s6 services are assembled in this Dockerfile.
• latest and cuda are Dockerfile targets, not separate branches.
• s6-overlay bits live under root/.
• Renovate tracks the upstream Obico commit from the bake annotations.