Woosuk Kwon 213f681f81 Revert "[Perf][GLM-5.2] Blackwell decode optimizations" (#49768)

2026-07-24 16:04:59 -07:00

..

attention

[Bugfix][Attention] Ignore empty MLA context chunks during merge (#49294)

2026-07-22 02:23:48 +00:00

core

[Refactor] Move fla to third party (#48500)

2026-07-16 19:22:36 +01:00

helion

[Kernel][Helion][1/N] Add Helion kernel for silu_and_mul_per_block_quant (#43994)

2026-07-07 00:19:01 +08:00

ir

[CI] Use explicit devices in IR tests (#49513)

2026-07-23 20:08:14 -07:00

mamba

[Kernel] ReplaySSM: cache SSM inputs for faster Mamba2 standard decode (#48018)

2026-07-24 09:39:49 -07:00

moe

[WideEP] Update NCCL to 2.30.7 to enable DeepEPv2 in the vllm/vllm-openai image (#45321)

2026-07-24 13:00:02 -07:00

quantization

[CI] Use explicit devices in quantization tests (#49512)

2026-07-23 10:54:26 -06:00

__init__.py

…

allclose_default.py

…

conftest.py

[CI] Fix stale/fragile untethered kernels-root tests (#49423)

2026-07-22 14:43:07 -06:00

quant_utils.py

…

test_apply_repetition_penalties.py

…

test_awq_int4_to_int8.py

…

test_bf16x3_router_gemm_cutedsl.py

BF16x3 router GEMM (#47973)

2026-07-16 17:04:20 +08:00

test_cache_kernels.py

[Bugfix] Restore gather_and_maybe_dequant_cache OOB guard (#49427)

2026-07-22 13:47:37 -07:00

test_compressor_kv_cache.py

[DSv4 Perf] Skip empty c128 kernel launch, around 2x kernel performance improvement. (#48957)

2026-07-22 10:55:20 -04:00

test_concat_mla_q.py

…

test_cp_gather_fp8.py

[Bugfix] Sparse MLA: enable fp8_ds_mla dense prefill (#48642)

2026-07-16 22:44:04 +00:00

test_fla_layernorm_guard.py

[Refactor] Move fla to third party (#48500)

2026-07-16 19:22:36 +01:00

test_flex_attention.py

[CI] Fix stale/fragile untethered kernels-root tests (#49423)

2026-07-22 14:43:07 -06:00

test_fp32_router_gemm.py

FP32 router GEMV optimization (#48335)

2026-07-11 13:07:48 +00:00

test_fused_deepseek_v4_qnorm_rope_kv_insert.py

[ROCm][CI] Relax fused layernorm quant test tolerances for one-ULP outliers (#46658)

2026-06-27 10:44:29 +08:00

test_fused_deepseek_v32_norm_rope.py

Revert "[Perf][GLM-5.2] Blackwell decode optimizations" (#49768)

2026-07-24 16:04:59 -07:00

test_fused_gdn_post_conv.py

[Refactor] Move fla to third party (#48500)

2026-07-16 19:22:36 +01:00

test_fused_indexer_q_rope_quant.py

…

test_fused_inv_rope_fp8_quant.py

[CI] Fix stale/fragile untethered kernels-root tests (#49423)

2026-07-22 14:43:07 -06:00

test_fused_minimax_m3_qknorm_rope_kv_insert.py

[CI] Fix stale/fragile untethered kernels-root tests (#49423)

2026-07-22 14:43:07 -06:00

test_fused_qk_norm_rope_gate.py

[Perf] fuse qk rmsnorm rope gate for qwen3.5 (#44176)

2026-06-09 22:12:17 +08:00

test_fused_quant_activation.py

…

test_fused_recurrent_packed_decode.py

[CI] Fix stale/fragile untethered kernels-root tests (#49423)

2026-07-22 14:43:07 -06:00

test_fused_sigmoid_gating_delta_rule.py

[CI] Fix stale/fragile untethered kernels-root tests (#49423)

2026-07-22 14:43:07 -06:00

test_kda.py

[Refactor] Move fla to third party (#48500)

2026-07-16 19:22:36 +01:00

test_ll_bf16_gemm.py

[Model] Add Inkling model support [1/N] (#48799)

2026-07-15 23:40:07 -07:00

test_mhc_kernels.py

[ROCm][DSV4] Disable TileLang MHC dispatch on gfx942 (#45931)

2026-06-22 09:26:54 +00:00

test_minimax_m3_amd_ops.py

[ROCm] fixed aiter master flag and expert parallelism compatibility on minimax-m3-mxfp8 (#47158)

2026-07-08 15:26:17 +08:00

test_minimax_m3_sparse_attn_fp8_scale.py

[ROCm][MiniMax-M3] Add AITER sparse paged attention (#47287)

2026-07-12 19:27:29 -07:00

test_onednn.py

…

test_shuffle_rows.py

…

test_top_k_per_row.py

[Perf][DSv4/DSv3.2] Add cluster-cooperative topK kernel for low-latency scenarios (#43008)

2026-06-23 16:11:00 -07:00

utils.py

[ROCm][CI] Relax fused layernorm quant test tolerances for one-ULP outliers (#46658)

2026-06-27 10:44:29 +08:00