Repository Issue Activity (beta feature)

#

#2670 linear_quantize_activations leaks ~1 temp .mlpackage per calibration op-group (tens of GB on realistic models)

#

#2646 Support Python 3.14

#

#

#2599 [PyTorch Conversion] SmolVLM model fails due to unsupported 'unfold' op in Core ML

#

#

#

#

#

#

#

#

#

#2566 Is it impossible to accelerate the decode of the Transformer using an NPU

#

#

#

#2667 InstanceNorm3d crashes on CoreML GPU path but works on CPU path

#

#2666 CoreML CPU InstanceNorm3d behaves differently depending on track_running_stats

#

#