Re:Show HN: FP32 matmul of large matrices up to 24% faster t...
by ap4 on Wednesday July 31, 2024 @08:52PM
(#41125169)
I did more tests on various GPUs. The TFLOPS values for A100 exceed the maximum from the spec of the GPU, so perhaps TFLOPS is calculated in a different way in the spec.
[ Parent
| on HN ]
There are no comments on this comment yet.