Skip to content

Performance Regression: PTQ-INT8 Quantization for nn.Conv2d with parameter groups > 1 is Slower than FP16 #134

Performance Regression: PTQ-INT8 Quantization for nn.Conv2d with parameter groups > 1 is Slower than FP16

Performance Regression: PTQ-INT8 Quantization for nn.Conv2d with parameter groups > 1 is Slower than FP16 #134

Triggered via issue July 31, 2025 03:44
@nvpohanhnvpohanh
commented on #4529 b8db91e
Status Skipped
Total duration 1s
Artifacts

feedback-update.yml

on: issue_comment
remove_label
0s
remove_label
Fit to window
Zoom out
Zoom in