Why the time cost of inferring resnet50-v2-7 with tvm-cuda much smallar than using tensorrt on x86? #15276
Labels
needs-triage
PRs or issues that need to be investigated by maintainers to find the right assignees to address it
type: bug
It cost 0.11ms when I inferred resnet50-v2-7 with tvm-cuda on x86.
Then I used the tensorrt to infer the same model with fp16, it cost 0.4ms .
Is it normal that tvm-cuda so much faster than tensorrt on x86?
The text was updated successfully, but these errors were encountered: