# NVIDIA GeForce RTX 4070 Ti, 60 SMs, sm_89, iters=20000 reps=5
mode       best TF/s   all runs
f16/f16        169.4 156.7/158.4/169.4/169.4/169.4
f16/f32         84.8 84.8/84.8/84.8/84.8/84.8
bf16/f32        84.7 84.7/84.7/84.7/84.7/84.7
s8/s32         338.8 338.8/338.8/338.0/338.0/338.0
dp4a            84.1 84.0/84.1/84.1/84.0/83.9
ffma            40.1 40.1/40.1/40.1/40.1/40.1
