Hi,
About the following question:https://developer.qualcomm.com/forum/qdn-forums/software/qualcomm-neural-processing-sdk/70540
I have more detailed test data for you:
API inference-time memory-usage cpu-usage
C++ API 52ms 260M 12%
C API 74ms 250M 7%
I hope it can help you analyze the problem. By the way, can the next version update fix the gap in C API time?
BR.
Dear developer,
Thanks for your sharing. We will discuss internal for this issue and get back to you later.
BR.
Wei