ODS V3 Pre-Release: Public testing and refinement ahead of the official V3 launch. Turn your PC, Mac, or Linux box into a private AI server.
Python ai-agentsamdcomfyuidockerllama-cppllmlocal-ain8nnvidiaopen-webuiragself-hostedspeech-to-textstrix-halotext-to-speechworkflow-automation
# ComfyUI — Image Generation
#
# This stub exists so the service registry can detect ComfyUI as installed.
# The actual container definition lives in GPU-specific overlay files that are
# merged with this stub at runtime by the installer/compose resolver:
#
# compose.amd.yaml — AMD ROCm (single GPU, native gfx architecture)
# compose.nvidia.yaml — NVIDIA CUDA (single GPU, builds local Dockerfile)
# compose.multigpu-amd.yaml — AMD ROCm multi-GPU
# compose.multigpu-nvidia.yaml — NVIDIA multi-GPU
#
# The correct overlay is selected by the installer based on the detected GPU backend.
# Do NOT add a service definition here; add it to the appropriate overlay file.
services: {}