Ma Mingfei f90b400431 [CPU] add support for mamba causal conv1d for qwen3-next (#12309)

2025-12-04 13:41:42 +08:00

..

mamba

[CPU] add support for mamba causal conv1d for qwen3-next (#12309)

2025-12-04 13:41:42 +08:00

model

[CPU] add fused_qkvzba_split_reshape_cat kernel for Qwen3-next (#12330)

2025-12-03 23:46:08 +08:00

activation.cpp

…

bmm.cpp

…

CMakeLists.txt

[CPU] Support chunk_gated_delta_rule kernel for Qwen3-Next (#12441)

2025-12-03 17:03:48 +08:00

common.h

[CPU] add support for mamba causal conv1d for qwen3-next (#12309)

2025-12-04 13:41:42 +08:00

decode.cpp

…

extend.cpp

[CPU] Support chunk_gated_delta_rule kernel for Qwen3-Next (#12441)

2025-12-03 17:03:48 +08:00

gemm_fp8.cpp

use faster covnersion from float8_e4m3fn to bfloat16 (#12316)

2025-12-04 12:34:05 +08:00

gemm_int8.cpp

…

gemm.cpp

…

gemm.h

…

interface.cpp

…

moe_fp8.cpp

…

moe_int8.cpp

…

moe.cpp

…

norm.cpp

Add fused_rmsnorm_gated_cpu kernel for CPU to support Qwen3-Next (#11577)

2025-11-21 01:33:31 +08:00

numa_utils.cpp

…

qkv_proj.cpp

…

rope.cpp

…

shm.cpp

…

shm.h

…

topk.cpp

…

torch_extension_cpu.cpp

[CPU] add support for mamba causal conv1d for qwen3-next (#12309)

2025-12-04 13:41:42 +08:00

vec_pack.h

[CPU] Support chunk_gated_delta_rule kernel for Qwen3-Next (#12441)

2025-12-03 17:03:48 +08:00

vec.h

use faster covnersion from float8_e4m3fn to bfloat16 (#12316)

2025-12-04 12:34:05 +08:00