Skip to content

Develop#143

Merged
solderzzc merged 4 commits intomasterfrom
develop
Mar 10, 2026
Merged

Develop#143
solderzzc merged 4 commits intomasterfrom
develop

Conversation

@solderzzc
Copy link
Member

No description provided.

solderzzc and others added 4 commits March 9, 2026 23:19
On Apple Silicon, force CoreML compute_units to CPU_AND_NE so object
detection runs on the 16-core Neural Engine instead of GPU, leaving
GPU free for LLM/VLM inference.

Changes:
- env_config.py: add compute_units field to BackendSpec, MPS defaults
  to cpu_and_ne, monkey-patch coremltools.MLModel during YOLO load
- detect.py: report compute_units in ready event
- SKILL.md: add compute_units parameter, update acceleration table
- test_env_config_ane.py: 15 unit tests for NPU logic
feat(detection): route Apple Silicon YOLO to Neural Engine (NPU)
…deploy.sh

- yolo-detection-2026: full GPU detection, venv, requirements, model optimization
- homesafe-bench: npm install
- smarthome-bench: yt-dlp/ffmpeg checks + npm install

ROCm/MPS blocks omitted (Linux/macOS only). Uses py.exe launcher
for Python discovery with fallback to python/python3 on PATH.
@solderzzc solderzzc merged commit 314242d into master Mar 10, 2026
1 check passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant