Wentao Ye ffb2cd6b54

[Perf] Optimize moe_align_block_size CUDA kernel (#19572) ...

Signed-off-by: yewentao256 <zhyanwentao@126.com>
Co-authored-by: mgoin <mgoin64@gmail.com>

2025-06-17 11:49:26 -07:00

..

attention

[Kernel] Raise verbose error and consolidate num_heads/num_kv_heads divisibility check (#19339)

2025-06-15 13:43:48 +08:00

core

…

mamba

…

moe

[Perf] Optimize moe_align_block_size CUDA kernel (#19572)

2025-06-17 11:49:26 -07:00

quantization

[Perf] Vectorize static / dynamic INT8 quant kernels (#19233)

2025-06-12 06:51:41 -07:00

__init__.py

…

allclose_default.py

…

quant_utils.py

…

test_apply_repetition_penalties.py

…

test_cutlass_mla_decode.py

…

test_flex_attention.py

Fixes IMA for TP w/ flex-attention (#19712)

2025-06-17 04:01:50 +00:00

test_fused_quant_activation.py

…

test_triton_flash_attention.py

…

utils.py

…