wozeparrotandGitHub c9b60caf8c gptoss: moe gemm kernels (#17178)

2026-07-24 07:25:15 -07:00

..

include

fa: support D=64 and attn sink (#16935)

2026-07-08 20:03:47 -07:00

fa_bwd_causal.cpp

llama: gemm/fa backward speedups (gpt 5.6) (#17007)

2026-07-14 00:27:00 +09:00

fa_bwd_post.cpp

fa: support D=64 and attn sink (#16935)

2026-07-08 20:03:47 -07:00

fa_bwd_pre.cpp

fa: support D=64 and attn sink (#16935)

2026-07-08 20:03:47 -07:00

fa_fwd_causal.cpp

fa: support D=64 and attn sink (#16935)

2026-07-08 20:03:47 -07:00

fa.py

llama: correct fused qkv shape assert (#17086)

2026-07-20 15:51:21 +09:00

fused_qkv_rope_bwd.cpp

fused qkv rope custom kernel (#17021)

2026-07-15 01:08:42 +09:00

fused_qkv_rope.cpp

fused qkv rope custom kernel (#17021)

2026-07-15 01:08:42 +09:00

gemm_bf16_atb.cpp

llama: a_bT and aT_b bf16 gemms (#16487)

2026-06-04 23:30:21 +09:00

gemm_bf16.cpp

llama: fix bf16 gemm oob (#16603)

2026-06-12 19:43:05 -07:00

gemm_fp8_atb.cpp

llama: gemm/fa backward speedups (gpt 5.6) (#17007)

2026-07-14 00:27:00 +09:00

gemm_fp8.cpp

llama: gemm/fa backward speedups (gpt 5.6) (#17007)

2026-07-14 00:27:00 +09:00

gemm_mxfp8.cpp

gemm: fix mxfp8 on more shapes (#16677)

2026-06-19 13:28:53 -07:00

grouped_mxfp8_gemm.cpp

gptoss: moe gemm kernels (#17178)

2026-07-24 07:25:15 -07:00

grouped_mxfp8_wgrad.cpp

gptoss: moe gemm kernels (#17178)

2026-07-24 07:25:15 -07:00