# NVIDIA GeForce RTX 4060 Ti, 34 SMs, sm_89, iters=20000 reps=5
mode       best TF/s   all runs
f16/f16         93.1 88.8/93.1/92.8/92.7/92.8
f16/f32         92.8 92.8/92.7/92.7/92.7/92.7
bf16/f32        92.7 92.7/92.3/92.2/92.2/92.2
s8/s32         184.5 184.4/184.5/184.4/184.4/184.5
dp4a            46.0 46.0/46.0/46.0/45.9/45.9
ffma            22.2 22.1/22.2/22.2/22.1/22.1
