-LC0 metal improvements
-core-ml
-fp16 working with metal and core-ml
-additional usage of Apple M5 Tensor cores
-usage of GPU + Apple Neural Engine (NPU) at the same time
= That's at least 100 times faster, than it is now
Moderator: Ras
Hai wrote: ↑Mon Nov 17, 2025 11:37 am The developers are working on:
-LC0 metal improvements![]()
-core-ml![]()
-fp16 working with metal and core-ml![]()
-additional usage of Apple M5 Tensor cores![]()
-usage of GPU + Apple Neural Engine (NPU) at the same time![]()
= That's at least 100 times faster, than it is now
Help is welcome
![]()
Did you actually read anything the devs said? Very bold claims when you have no evidenceHai wrote: ↑Mon Nov 17, 2025 11:37 am The developers are working on:
-LC0 metal improvements![]()
-core-ml![]()
-fp16 working with metal and core-ml![]()
-additional usage of Apple M5 Tensor cores![]()
-usage of GPU + Apple Neural Engine (NPU) at the same time![]()
= That's at least 100 times faster, than it is now
Help is welcome
![]()