A rather brief story about overwhelming demand for Nvidia's high-performance computing hardware in China has revealed the performance of Nvidia's mysterious A800 compute GPU, which is made for the Chinese market. According to MyDrivers, the A800 operates at 70% of the speed of A100 GPUs while complying with strict U.S. export standards that limit how much processing power Nvidia can sell.
Being three years old now, Nvidia's A100 is quite a performer: it delivers 9.7 FP64/19.5 FP64 Tensor TFLOPS for HPC and up to 624 BF16/FP16 TFLOPS (with sparsity) for AI workloads. Even being cut by around 30%, these numbers will still look formidable: 6.8 FP64/13.7 FP64 Tensor TFLOPS as well as 437 BF16/FP16 (with sparsity).
