wozeparrotandGitHub 05249466ed llama: fused quantize mxfp8 (#16667)

2026-06-18 16:02:28 -07:00

..

cast_amax

fp8 gemm inv_scale in epilogue (#16625)

2026-06-15 18:44:41 +09:00

fp8_transpose

llama speed 6 (#16071)

2026-05-06 20:51:03 -07:00

fused_ce

llama: no E_ copy after bf16 GEMM (#16458)

2026-06-02 14:14:13 +09:00

fused_rmsnorm_mul_quantize_fp8

fp8 gemm inv_scale in epilogue (#16625)

2026-06-15 18:44:41 +09:00

quantize_fp8_delayed

quantize_fp8 kernels in uops (#16288)

2026-05-22 20:54:06 +09:00

quantize_mxfp8_fused

llama: fused quantize mxfp8 (#16667)

2026-06-18 16:02:28 -07:00

rmsnorm

llama: move llama kernels to llama_kernels (#15952)

2026-04-27 22:48:53 -07:00

__init__.py

llama: update local amax implementation after ParamArgs change (#16446)

2026-05-30 16:55:43 +09:00