That's strange because the cutlass docs explicitly does NOT mention fp8 support....

rurban on July 11, 2025 | parent | context | favorite | on: FP8 is ~100 tflops faster when the kernel name has...

That's strange because the cutlass docs explicitly does NOT mention fp8 support. So it looks like it can be used nevertheless with fp8 by using the name hack.

mlazos on July 11, 2025 [–]

It supports e5m2 and e4m3 right in the doc linked.