docker

Docker

Build the local development image from the repository root:

docker compose -f docker/docker-compose.yml build dev

Run a one-off CLI command:

docker compose -f docker/docker-compose.yml run --rm dev armor --help

Start the daemon inside the container with the baked Qwen3 GGUF model:

docker compose -f docker/docker-compose.yml run --rm dev \
  armor daemon --socket /tmp/armor.sock --db /tmp/armor.db --model /models/active.gguf

The Dockerfile downloads both model artifacts during the builder stage:

• lmstudio-community/Qwen3-0.6B-GGUF / Qwen3-0.6B-Q4_K_M.gguf
• sentence-transformers/all-MiniLM-L6-v2 / onnx/model.onnx

No Hugging Face token is required for these public models. Unauthenticated builds may print a rate-limit warning; set HF_TOKEN in the Docker build environment only if your network needs higher Hugging Face rate limits.

Last verified on 2026-05-09: docker compose -f docker/docker-compose.yml build --no-cache dev usually completes in under 3 minutes on the README benchmark host and produces armor-dev at about 990 MiB.