[Attention][MiniMax-M3] NVFP4 KV cache on the MSA sparse attention path (#59300)

Signed-off-by: Yongye Zhu <zyy1102000@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
Yongye Zhu
2026-10-01 14:39:55 -07:00
committed by GitHub
co-authored by Claude Opus 5.5
parent ea84051819
commit 7566d83bd3
12 changed files with 1297 additions and 73 deletions
+21
View File
@@ -590,9 +590,30 @@ if(VLLM_GPU_LANG STREQUAL "CUDA" OR VLLM_GPU_LANG STREQUAL "HIP")
endif()
endif()
# The fused MiniMax-M3 KV insert quantizes NVFP4 with SM100-family-only
# cvt instructions, so its SM10x/11x code is built for the family targets.
set(MINIMAX_M3_FUSED_SRC
"csrc/libtorch_stable/fused_minimax_m3_qknorm_rope_kv_insert_kernel.cu")
list(REMOVE_ITEM VLLM_STABLE_EXT_SRC "${MINIMAX_M3_FUSED_SRC}")
set_gencode_flags_for_srcs(
SRCS "${VLLM_STABLE_EXT_SRC}"
CUDA_ARCHS "${CUDA_ARCHS}")
if(${CMAKE_CUDA_COMPILER_VERSION} VERSION_GREATER_EQUAL 13.0)
cuda_archs_loose_intersection(MINIMAX_M3_FUSED_ARCHS
"10.0f;10.7f;11.0f" "${CUDA_ARCHS}")
else()
cuda_archs_loose_intersection(MINIMAX_M3_FUSED_ARCHS
"10.0a;10.1a;10.3a" "${CUDA_ARCHS}")
endif()
foreach(_ARCH ${CUDA_ARCHS})
if(_ARCH VERSION_LESS 10.0 OR _ARCH VERSION_GREATER_EQUAL 12.0)
list(APPEND MINIMAX_M3_FUSED_ARCHS "${_ARCH}")
endif()
endforeach()
set_gencode_flags_for_srcs(
SRCS "${MINIMAX_M3_FUSED_SRC}"
CUDA_ARCHS "${MINIMAX_M3_FUSED_ARCHS}")
list(APPEND VLLM_STABLE_EXT_SRC "${MINIMAX_M3_FUSED_SRC}")
if(COOPERATIVE_TOPK_ARCHS)
list(APPEND VLLM_STABLE_EXT_SRC