Description
Deep learning inference can use Intel NPU acceleration through an OpenVINO plugin. This helps developers run supported models on compatible neural processing hardware.
Accelerator plugins depend on driver, runtime, and model compatibility. Validate output quality and hardware support before deployment.