# NVIDIA GeForce RTX 4090, 128 SMs, sm_89, iters=20000 reps=5
mode       best TF/s   all runs
f16/f16        330.8 302.7/302.7/314.8/330.8/328.9
f16/f32        328.9 328.9/328.9/328.9/328.9/328.9
bf16/f32       328.8 328.8/327.6/326.3/326.6/327.3
s8/s32         657.6 652.9/657.6/657.6/657.6/657.6
dp4a           163.4 163.4/163.4/163.4/163.4/163.4
ffma            78.5 78.5/78.5/78.5/78.5/77.8
