About the Maximum Achievable Matmul FLOPS comparison table

In the Maximum Achievable Matmul FLOPS comparison table, the theory value of NVIDIA H100 SXM is 989. However, in the NVIDIA H100 Tensor Core GPU Datasheet, the value of BFLOAT16 Tensor Core is 1,979. Is there a mistake in my understanding here? - https://github.com/stas00/ml-engineering/tree/master/compute/accelerator#maximum-achievable-matmul-flops-comparison-table - https://resources.nvidia.com/en-us-tensor-core/nvidia-tensor-core-gpu-datasheet