zlacker

>>rafael+(OP)
It is a bug in MLX that has been fixed a few days ago: https://github.com/ml-explore/mlx/pull/3083

>>zcbenz+Ys1
So the underlying issue is that the iPhone 16 Pro SKU was misdetected as having Neural Accelerator (nax) support and this caused silently wrong results. Not a problem with the actual hardware.

>>zozbot+3v1
From a debugging point of view, the author's conclusion was still completely reasonable given the evidence they had

>>TimByt+7O1
No it wasn't. A hardware defect so disastrous that it affects floating point computation on the neural engine, yet so minor that it does not affect any of the software on the device utilizing that hardware is exceedingly improbable.

The conclusion, that it was not the fault of the developer was correct, but assuming anything other than a problem at some point in the software stack is unreasonable.