vasanthzoh

~Thread spinning logic (for `force_spinning_stop_between_runs_`) is managed only in the case of per-session threads based on the code [here](https://github.com/microsoft/onnxruntime/blob/main/onnxruntime/core/session/inference_session.cc#L2954-L2960). But WASM builds are forced to use global threads.~ 1. ~Could...

Hi @fs-eire @snnn, I'm following up on this issue. I can confirm that I similarly face much higher latencies and CPU usage when doing inference on WASM as compared to...

@fs-eire I'm not sure which code you are referring to, could you please elaborate?

Thanks for the clarification @shriramrsr Reiterating my original point once again as our confusion is now cleared @fs-eire I can confirm that I similarly face much higher latencies and CPU...