1 comment

[ 1.4 ms ] story [ 10.7 ms ] thread
Think AI did this? Finding optimizations seems to be happening everywhere. Even llama.cpp is getting faster using AI, which they dont accept AI patches, so forks everywhere.