Kebe 5ee5c86eeb [Bugfix][DeepSeek-V3.2] fix fp8 kvcache type cast (#33884) ...

Signed-off-by: Kebe <mail@kebe7jun.com>

2026-02-10 19:31:36 -08:00

..

attention

[Performance][MLA][ROCm] Remove redundant D2D copy in deepseek (#27457)

2025-11-26 12:45:28 +08:00

core

…

cpu

[Fix] [CPU Backend] : Prepack weights for w8a8 oneDNN matmul (#33901)

2026-02-09 18:04:41 +08:00

cutlass_extensions

[Kernel] Add enable_sm120_or_later for SM121 (DGX Spark) CUTLASS support (#33517)

2026-02-06 20:28:01 -08:00

mamba/mamba_ssm

…

moe

[Refactor] Remove align block size logic in moe_permute (#33449)

2026-02-06 10:57:06 -08:00

quantization

[Kernel] Add enable_sm120_or_later for SM121 (DGX Spark) CUTLASS support (#33517)

2026-02-06 20:28:01 -08:00

quickreduce

…

rocm

Perf tuning and expansion of cases covered for wvSplitKrc (#33493)

2026-02-07 05:33:11 -08:00

sparse/cutlass

…

activation_kernels.cu

[Perf][Kernels] Vectorize csrc/activations_kernels.cu (#29512)

2025-12-16 14:56:02 -08:00

cache_kernels_fused.cu

Fuse RoPE and MLA KV-cache write (#25774)

2026-01-09 19:18:37 -08:00

cache_kernels.cu

[Bugfix][DeepSeek-V3.2] fix fp8 kvcache type cast (#33884)

2026-02-10 19:31:36 -08:00

cache.h

OffloadingConnector: Support kernel_block_size != block_size (#30692)

2026-01-22 12:30:04 +00:00

cub_helpers.h

…

cuda_compat.h

…

cuda_utils_kernels.cu

…

cuda_utils.h

…

cuda_view.cu

Simplify from_blob usage in get_cuda_view_from_cpu_tensor (#29027)

2025-11-22 10:35:32 +00:00

cumem_allocator_compat.h

…

cumem_allocator.cpp

[Feature][Core] Support Fabric detection to adapt the MNNVL protocol for the GB series (#33540)

2026-02-02 22:55:46 +08:00

custom_all_reduce_test.cu

…

custom_all_reduce.cu

…

custom_all_reduce.cuh

…

custom_quickreduce.cu

…

dispatch_utils.h

[Performance] Fused blockwise quant RMS norm (#27883)

2025-12-07 16:38:04 +00:00

fused_qknorm_rope_kernel.cu

[ROCm] [Critical]: Remove unused variable (#31156)

2025-12-22 08:28:22 -08:00

launch_bounds_utils.h

…

layernorm_kernels.cu

[kernel][perf] support uncontiguous input for rms_norm kernel (#28103)

2025-11-20 19:39:09 -08:00

layernorm_quant_kernels.cu

…

ops.h

[Perf][Kernel] Add faster topKperRow decode kernel for DeepSeek-V3.2 sparse attention (#33680)

2026-02-10 10:29:52 -05:00

permute_cols.cu

…

pos_encoding_kernels.cu

…

sampler.cu

[Perf][Kernel] Add faster topKperRow decode kernel for DeepSeek-V3.2 sparse attention (#33680)

2026-02-10 10:29:52 -05:00

topk.cu

[Perf][Kernel] Add faster topKperRow decode kernel for DeepSeek-V3.2 sparse attention (#33680)

2026-02-10 10:29:52 -05:00

torch_bindings.cpp

[Perf][Kernel] Add faster topKperRow decode kernel for DeepSeek-V3.2 sparse attention (#33680)

2026-02-10 10:29:52 -05:00

type_convert.cuh

…