3 people c8cfb9400a Fix multi-GPU OpenVINO detection for enrichments (#23188) ...

On multi-GPU systems, OpenVINO enumerates devices as "GPU.0", "GPU.1",
etc. rather than a single "GPU". The exact string match in
is_openvino_gpu_npu_available() fails to recognize these suffixed device
names, causing enrichments (face recognition, semantic search) to
silently fall back to CPU-only inference via ONNXModelRunner instead of
using OpenVINOModelRunner on GPU.

Switch from exact match to prefix match so both single-GPU ("GPU") and
multi-GPU ("GPU.0", "GPU.1") device names are correctly detected, along
with any future suffixed variants for NPU and other accelerators.

Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

2026-05-13 11:22:55 -06:00

..

plugins

Miscellaneous fixes (#23082)

2026-05-01 11:25:26 -06:00

__init__.py

Add isort and ruff linter (#6575)

2023-05-29 05:31:17 -05:00

detection_api.py

Implement YOLOx for RKNN (#17788)

2025-04-18 14:44:02 -06:00

detection_runners.py

Fix multi-GPU OpenVINO detection for enrichments (#23188)

2026-05-13 11:22:55 -06:00

detector_config.py

Tweaks (#22656)

2026-03-26 13:47:24 -05:00

detector_types.py

Add isort and ruff linter (#6575)

2023-05-29 05:31:17 -05:00

detector_utils.py

Fallback from tflite-runtime to ai-edge-litert (#21876)

2026-02-26 21:55:29 -07:00