Woosuk Kwon 752c6ade2e

[V0 Deprecation] Deprecate BlockSparse Attention & Phi3-Small (#21217) ...

Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>

2025-07-19 13:53:17 -07:00

..

attention

[V0 Deprecation] Deprecate BlockSparse Attention & Phi3-Small (#21217)

2025-07-19 13:53:17 -07:00

core

[CI] change spell checker from codespell to typos (#18711)

2025-06-11 19:57:10 -07:00

mamba

[Kernel] Triton implementation of causal-conv1d for Mamba-based models (#18218)

2025-07-09 12:53:55 -07:00

moe

Add torch golden impl for moe_align_block_size kernel test (#20653)

2025-07-19 02:32:36 -07:00

quantization

[Perf] Use Triton instead of Torch for DeepGEMM Per Token Group Quant (#20841)

2025-07-12 19:38:45 -07:00

__init__.py

…

allclose_default.py

…

quant_utils.py

[Kernel] Enable fp8 support for pplx and BatchedTritonExperts. (#18864)

2025-07-03 14:55:40 -07:00

test_apply_repetition_penalties.py

[BUG] Fix #20484. Support empty sequence in cuda penalty kernel (#20491)

2025-07-05 19:38:02 -07:00

test_cutlass_mla_decode.py

…

test_flex_attention.py

[Misc] Add SPDX-FileCopyrightText (#20428)

2025-07-04 07:40:42 +00:00

test_fused_quant_activation.py

…

test_triton_flash_attention.py

…

utils.py

[Misc] Add unit tests for MoE ModularKernel combinations + Profiling utility (#20449)

2025-07-11 07:51:46 -07:00