Slow inference speed after my core ml model was encrypted

Question

Created Apr ’25

Replies 4

Boosts 0

Participants 2

Hi friends,

I have just found that the inference speed dropped to only 1/10 of the original model.

Had anyone encountered this?

Thank you.

Answered by DTS Engineer in 835046022

Please file a bug report for this issue using Feedback Assistant. It is unexpected that model encryption would affect inference time.

-- Greg

Boost

Answer 1

Apr ’25

I profiled the app in 'Instruments', found that it was using the CPU to inference, so it is much slower than the original model.

1

Answer 2

DTS Engineer OP

Apple

Apr ’25

Recommended

Please file a bug report for this issue using Feedback Assistant. It is unexpected that model encryption would affect inference time.

-- Greg

0

Answer 3

Apr ’25

Fired the bug.

0

Answer 4

Apr ’25

Filed the bug with 'Feedback Assistant'.

0