# NVIDIA GeForce RTX 4070 Ti, 60 SMs, sm_89, iters=20000 reps=5
mode       best TF/s   all runs
f16/f16        170.0 158.6/170.0/169.4/169.4/169.4
f16/f32        169.4 169.4/169.4/169.4/169.4/169.4
bf16/f32       168.9 168.7/168.9/168.9/168.9/168.9
s8/s32         338.0 337.9/337.9/338.0/338.0/337.9
dp4a            84.0 84.0/84.0/84.0/84.0/84.0
ffma            40.2 40.2/40.2/40.1/40.1/40.1
