8 ms·That's why you don't run the model at low quantizationby rescbr 2mo agoThat's why you don't run the model at low quantization