[11/05/25 21:55:21] INFO Starting gunicorn 23.0.0 [11/05/25 21:55:21] INFO Listening at: http://[::]:3003 (8) [11/05/25 21:55:21] INFO Using worker: immich_ml.config.CustomUvicornWorker [11/05/25 21:55:21] INFO Booting worker with pid: 9 [11/05/25 21:55:23] INFO Started server process [9] [11/05/25 21:55:23] INFO Waiting for application startup. [11/05/25 21:55:23] INFO Created in-memory cache with unloading after 300s of inactivity. [11/05/25 21:55:23] INFO Initialized request thread pool with 16 threads. [11/05/25 21:55:23] INFO Application startup complete. [11/05/25 21:57:54] INFO Loading detection model 'antelopev2' to memory [11/05/25 21:57:54] INFO Setting execution providers to ['OpenVINOExecutionProvider', 'CPUExecutionProvider'], in descending order of preference 2025-11-05 21:57:54.994176143 [W:onnxruntime:Default, openvino_provider_factory.cc:260 operator()] Unsupported device key: GPU.0. Skipping entry. [11/05/25 21:57:56] INFO Loading visual model 'ViT-H-14-378-quickgelu__dfn5b' to memory [11/05/25 21:57:56] INFO Setting execution providers to ['OpenVINOExecutionProvider', 'CPUExecutionProvider'], in descending order of preference 2025-11-05 21:57:56.620088555 [W:onnxruntime:Default, openvino_provider_factory.cc:260 operator()] Unsupported device key: GPU.0. Skipping entry. [11/05/25 21:58:11] INFO Downloading detection model 'PP-OCRv5_mobile' to /cache/ocr/PP-OCRv5_mobile/detection/model.onnx. This may take a while. [11/05/25 21:58:11] INFO Initiating download: https://www.modelscope.cn/models/RapidAI/RapidOCR/r esolve/v3.4.0/onnx/PP-OCRv5/det/ch_PP-OCRv5_mobile_ det.onnx [11/05/25 21:58:14] INFO Download size: 4.60MB [11/05/25 21:58:14] INFO Successfully saved to: /cache/ocr/PP-OCRv5_mobile/detection/model.onnx [11/05/25 21:58:14] INFO Loading detection model 'PP-OCRv5_mobile' to memory [11/05/25 21:58:14] INFO Setting execution providers to ['OpenVINOExecutionProvider', 'CPUExecutionProvider'], in descending order of preference 2025-11-05 21:58:14.421447398 [W:onnxruntime:Default, openvino_provider_factory.cc:260 operator()] Unsupported device key: GPU.0. Skipping entry. [11/05/25 21:58:20] INFO Loading recognition model 'antelopev2' to memory [11/05/25 21:58:20] INFO Setting execution providers to ['OpenVINOExecutionProvider', 'CPUExecutionProvider'], in descending order of preference 2025-11-05 21:58:21.286469250 [W:onnxruntime:Default, openvino_provider_factory.cc:260 operator()] Unsupported device key: GPU.0. Skipping entry. [11/05/25 21:58:37] INFO Downloading recognition model 'PP-OCRv5_mobile' to /cache/ocr/PP-OCRv5_mobile/recognition/model.onnx. This may take a while. [11/05/25 21:58:37] INFO Initiating download: https://www.modelscope.cn/models/RapidAI/RapidOCR/r esolve/v3.4.0/onnx/PP-OCRv5/rec/ch_PP-OCRv5_rec_mob ile_infer.onnx [11/05/25 21:58:40] INFO Download size: 15.86MB [11/05/25 21:58:41] INFO Successfully saved to: /cache/ocr/PP-OCRv5_mobile/recognition/model.onnx [11/05/25 21:58:41] INFO Loading recognition model 'PP-OCRv5_mobile' to memory [11/05/25 21:58:41] INFO Setting execution providers to ['OpenVINOExecutionProvider', 'CPUExecutionProvider'], in descending order of preference 2025-11-05 21:58:41.217754298 [W:onnxruntime:Default, openvino_provider_factory.cc:260 operator()] Unsupported device key: GPU.0. Skipping entry. [INFO] 2025-11-05 21:58:43,378 [RapidOCR] base.py:22: Using engine_name: onnxruntime