I am new to TVM. Is one goal of TVM to accelerate the inference speed?

Inference speed on CPU is a rigid demand for users.

Thank you and thank you.

Just googled “TVM Bert cpu”.

Problem solved.