Robert Shaw 83a317f650

[MoE Refactor][3/N] Deprecate cutlass block quant fp8 (b200) (#30990) ...

Signed-off-by: Robert Shaw <robshaw@redhat.com>
Co-authored-by: Robert Shaw <robshaw@redhat.com>

2025-12-19 13:09:54 -08:00

..

attention

[MM Encoder]: Migrate legacy ViT MultiHeadAttention to new MMEncoderAttention interface (#30684)

2025-12-19 02:04:19 +08:00

core

[CustomOp] Extract ApplyRotaryEmb as CustomOp and unify the dispatch logic (#29873)

2025-12-15 19:08:16 -08:00

mamba

Add SpecDec support to selective_state_update (#29488)

2025-12-08 16:45:18 -05:00

moe

[MoE Refactor][3/N] Deprecate cutlass block quant fp8 (b200) (#30990)

2025-12-19 13:09:54 -08:00

quantization

[Bugfix] awq_gemm: fix argument order swap (#30364)

2025-12-14 18:15:37 +08:00

__init__.py

…

allclose_default.py

…

quant_utils.py

[CI/Build][AMD] Fix ref_dynamic_per_token_quant reference implementation on ROCm. (#30291)

2025-12-12 09:30:23 +00:00

test_apply_repetition_penalties.py

…

test_cache_kernels.py

[Bugfix][cache_kernels]: Fix OOB in cache_kernels.cu (#28760)

2025-11-20 02:52:02 -08:00

test_fla_layernorm_guard.py

…

test_flex_attention.py

[Fix][FlexAttention] return max logical block index to handle reused blocks (#30915)

2025-12-18 06:42:21 +00:00

test_fused_quant_activation.py

…

test_onednn.py

[CPU] Refactor CPU attention backend (#27954)

2025-11-12 09:43:06 +08:00

test_shuffle_rows.py

…

test_top_k_per_row.py

[DeepSeek v3.2] Make top-k work for any logit values. (#27568)

2025-12-08 06:55:58 -08:00

utils.py

[Feat] Support non-gated activations in NVFP4 modelopt path (#29004)

2025-11-30 11:02:40 -05:00