1
0
AlpinDale ba371fbbbd feat: AWQ marlin kernels (#603) 5 сар өмнө
..
amd 251568470e initial nvidia fp8 e4m3 for kv cache 6 сар өмнө
nvidia 3bdeb3e116 fix: clang formatting for all kernels (#558) 6 сар өмнө
common.cu c8f5424d72 add scale_ub inputs to fp8 dynamic per-token quant 5 сар өмнө
fp8_marlin.cu ba371fbbbd feat: AWQ marlin kernels (#603) 5 сар өмнө