COMMITS
/ benchmarks/mx_formats/cast_bench.py August 6, 2025
V
extend the MX cast benchmark to include casting to mxfp4 (#2693)
Vasiliy Kuznetsov committed
August 4, 2025
V
mx: expose scaling calculation methods in training UX (#2620)
Vasiliy Kuznetsov committed
July 15, 2025
D
Add CUDA kernel for MXFP8 dim1 casting (#2513)
Daniel Vega-Myhre committed
May 9, 2025
D
Add a triton kernel for swizziling (#2168)
Driss Guessous committed
April 1, 2025
V
mx: small speedup with dim0 cast (#1980)
Vasiliy Kuznetsov committed
March 26, 2025
V
mx: triton kernel to cast to mx and write in col-major (#1932)
Vasiliy Kuznetsov committed
March 18, 2025
V
mx cast performance bench (#1907)
Vasiliy Kuznetsov committed