Quartermaster
Local platform that auto-configures models for VRAM and serves chat, images, voice via API.
Tested on Linux
Use it
Linux x86_64
You need first
- Runtimes:
Go 1.26+ and Node.js 24 for the UI
download prebuilt quartermaster-linux-amd64 v1.7.2 and install to ~/.local/bin
rm -f "$HOME"/.local/bin/quartermaster && curl -fL -o "$HOME"/.local/bin/quartermaster https://github.com/Quartermaster-Labs/Quartermaster/releases/download/v1.7.2/quartermaster-linux-amd64 && chmod +x "$HOME"/.local/bin/quartermaster
Verified in pickfit: the commands exactly as its sandbox ran them
For the record only: they name the sandbox's home, /home/sbx.
rm -f /home/sbx/.local/bin/quartermaster && curl -fL -o /home/sbx/.local/bin/quartermaster https://github.com/Quartermaster-Labs/Quartermaster/releases/download/v1.7.2/quartermaster-linux-amd64 && chmod +x /home/sbx/.local/bin/quartermasterquartermaster -versioncat > /tmp/qm-test.yaml <<'YAML'
models:
dummy:
cmd: python3 -m http.server ${PORT}
YAML
cat /tmp/qm-test.yamlquartermaster -config /tmp/qm-test.yaml -no-generate -listen 127.0.0.1:18080curl -s http://127.0.0.1:18080/logsCheck that it works
quartermaster -versionThe output should contain “v1.7.2”.
First thing to try
cat > /tmp/qm-test.yaml <<'YAML'
models:
dummy:
cmd: python3 -m http.server ${PORT}
YAML
cat /tmp/qm-test.yamlquartermaster -config /tmp/qm-test.yaml -no-generate -listen 127.0.0.1:18080curl -s http://127.0.0.1:18080/logsRun quartermaster to start the local server, then open its address (such as http://127.0.0.1:18080) in your browser or point API clients at it.
What it does
- Chat, make images and use voice on your own machineNot tested
- Set up models automatically to fit your GPU memoryNot tested
- Switch models behind one shared APINot shown working
How far pickfit got
Linux x86_64
- Installs
- Starts
- 0 of 3 features work
- Works offline
macOS and Windows: the project says it runs there; pickfit hasn’t tried it.
Good for
- chat privately with local LLMs in household
- generate and edit images with local diffusion models
- synthesize speech and transcribe audio locally
- search Hugging Face and download quants from app
- serve local models via OpenAI-compatible API
How pickfit tested it
Each run starts from a clean machine and follows the steps above. How pickfit verifies
Linux x86_64
Its service answers requests
- Replayed from scratch
- Works offline
- Checked on 30 Oct 2026
- Version v1.7.2
Differences the test machine hit
- README: building from source requires Go 1.26+ and Node.js 24
In the test: used README-documented prebuilt quartermaster-linux-amd64 v1.7.2 verified against SHA256SUMS, no toolchain needed - README: smallest config shows cmd with ${PORT}
In the test: cmd must literally contain ${PORT}, e.g. python3 -m http.server ${PORT}, otherwise config load fails