Polynomial math lifts Blackwell LLM throughput by up to 8%
Replacing native sigmoid, tanh, and SiLU calculations with short polynomial programs can raise complete training-step throughput by up to 8.0% on GB200 systems, giving LLM builders a way to ease special-function bottlenecks without changing the model checkpoint. The method swaps each function for packed multiply-add arithmetic inside the kernel, with gains of 2.7% for dense SiLU, 2.9% for tanh-softcapped attention, and 8.0% for routed-expert SwiGLU. Across four tasks, final smoothed training-loss differences ranged from -0.107 to +0.079 near 100 billion tokens.