GPU
Edit@2 months ago
FP64: 1.13 TFLOPS
FP32: 36.13 TFLOPS
FP16: 72.25 TFLOPS
FP16: 144.51 TFLOPS, with sparsity: 289.01 TFLOPS
FP8: 289.01 TFLOPS, with sparsity: 578.03 TFLOPS
BF16: 144.51 TFLOPS, with sparsity: 289.01 TFLOPS
INT8: 289.01 TOPS, with sparsity: 578.03 TOPS
INT4: 578.03 TOPS, with sparsity: 1156.06 TOPS