Josh Casada a2c43ad8bb feat: ZMQ embedding runner for offloading ONNX inference to native host ... Extends the ZMQ split-detector pattern (apple-silicon-detector) to cover ONNX embedding models — ArcFace face recognition and Jina semantic search. On macOS, Docker has no access to CoreML or the Apple Neural Engine, so embedding inference is forced to CPU (~200ms/face for ArcFace). This adds a ZmqEmbeddingRunner that sends preprocessed tensors to a native host process over ZMQ TCP and receives embeddings back, enabling CoreML/ANE acceleration outside the container. Files changed: - frigate/detectors/detection_runners.py: add ZmqEmbeddingRunner class and hook into get_optimized_runner() via "zmq://" device prefix - tools/zmq_embedding_server.py: new host-side server script Tested on Mac Mini M4, 24h soak test, ~5000 object reindex.		2026-02-21 12:44:42 -05:00
..
zmq_embedding_server.py	feat: ZMQ embedding runner for offloading ONNX inference to native host	2026-02-21 12:44:42 -05:00