3 people ef43e3101b [ROCm][DSV4][Perf] Optimize Triton sparse-MLA decode on gfx950 (#52212) ...

Signed-off-by: Fangzhou Ai <fangzhou.ai@amd.com>
Signed-off-by: fai <fangzhouai@gmail.com>
Co-authored-by: Nick Hill <nickhill123@gmail.com>
Co-authored-by: Cursor Grok 4.6 <cursoragent@cursor.com>

2026-08-16 12:16:46 -07:00

..

attention

[ROCm][DSV4][Perf] Optimize Triton sparse-MLA decode on gfx950 (#52212)

2026-08-16 12:16:46 -07:00

core

[Bugfix] Fix .../mrope.py::apply_interleaved_rope() when torch.compile is used in torch==2.13 (#52005)

2026-08-13 22:08:40 +00:00

helion

[Kernel][Helion] Add numerics checks to benchmark script (#48968)

2026-07-31 20:16:02 -04:00

ir

[XPU] Unify XPU RMSNorm kernels with vllm_c and drop redundant XPU-specific implementation (#46981)

2026-07-31 17:28:18 +08:00

mamba

[Kernel][Perf] Add fused CUDA post-conv MTP decode kernel for Qwen3.5 GDN (#51674)

2026-08-14 11:44:45 +08:00

moe

[Bugfix] Make DSV4 sparse MLA work end-to-end for plain decode, MTP, and DSpark (#51538)

2026-08-15 07:01:32 -07:00

quantization

[Refactor] Remove dead code for quantization (#52221)

2026-08-16 13:10:16 -04:00

turboquant

[Kernel][ROCm][Perf] FlyDSL decode-attention kernel for 4-bit TurboQuant KV cache (#47896)

2026-08-11 08:45:39 -07:00

__init__.py

[CI/Build] Move test_utils.py to tests/utils.py (#4425)

2024-05-13 23:50:09 +09:00

allclose_default.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

conftest.py

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

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

quant_utils.py

[Perf] Create TMA-aligned input scale tensor for DeepGemm on Hopper (#32619)

2026-01-22 15:47:04 -05:00

test_apply_repetition_penalties.py

[Platform] Deprecate seed_everything (#31659)

2026-01-04 18:34:04 -08:00

test_awq_int4_to_int8.py

[CPU] Bump up to the latest CPU kernels (#41924)

2026-05-07 05:45:59 -07:00

test_bf16_skinny_gemm.py

[Kernel] Extend CuTe DSL skinny GEMM to GLM-5.2 (#49791)

2026-08-03 16:49:41 -07:00

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

[ROCm][DSV4][Perf] Optimize Triton sparse-MLA decode on gfx950 (#52212)

2026-08-16 12:16:46 -07:00

test_concat_mla_q.py

[Attention][Perf][Kernel] Replace torch.cat with vectorized CUDA kernel MLA query concat - DeepSeek-V3.2 (#34917)

2026-03-09 09:50:36 -07:00

test_cp_gather_fp8.py

[Kernel] Optimize long-context MLA cache gathers (#51739)

2026-08-10 22:23:43 -07:00

test_fla_layernorm_guard.py

[Misc] Add and enable Triton kernel unit tests on XPU (#45694)

2026-08-07 02:53:03 -07:00

test_flex_attention.py

[FlexAttention] Avoid encoder block-mask compile explosion (#50339)

2026-07-30 11:24:13 -04:00

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

[DSv4 Perf] Optimize workspace reuse for eager break, 3.9% E2E TTFT improvement. (#49236)

2026-07-31 08:45:58 -07:00

test_fused_deepseek_v32_norm_rope.py

[ROCm]Migrating Deepseek V3.2 to vllm/models/deepseek_v32/ (#47207)

2026-07-30 08:37:43 -05:00

test_fused_gdn_post_conv.py

[Kernel][Perf] Add fused CUDA post-conv MTP decode kernel for Qwen3.5 GDN (#51674)

2026-08-14 11:44:45 +08:00

test_fused_indexer_q_rope_quant.py

[DSv4 Perf] Optimize workspace reuse for eager break, 3.9% E2E TTFT improvement. (#49236)

2026-07-31 08:45:58 -07:00

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

[Attention][MiniMax-M3] Add MSA speculative decode verification (#50032)

2026-08-02 13:46:55 -07:00

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

Add platform manual_seed_all API (#38468)

2026-04-10 13:43:50 +08:00

test_fused_recurrent_packed_decode.py

[Bugfix] Fix packed GDN decode launch for large batch-head grids (#52030)

2026-08-13 07:19:08 -07:00

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_gate_linear_rocm_dispatch.py

[Hardware][AMD] Enable fused bf16→fp32 router GEMM on ROCm (#50268)

2026-08-12 02:46:13 +00:00

test_kimi_k3_gemm_rs.py

[Kimi-K3] Add GEMM-RS for sequence parallelism (#52079)

2026-08-13 09:02:28 -07:00

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_mhc_tilelang_jit.py

[ROCm] Defer tilelang import through its import from vllm.tilelang_utils import tilelang and relaxed has_tilelang (#51159)

2026-08-13 13:12:40 -05:00

test_minimax_m3_amd_ops.py

[ROCm][MoE] Fix expert_map vs AITER expert_mask for non-AITER experts under EP (#49758)

2026-08-11 15:33:02 -05:00

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

[CPU] Refactor CPU attention backend (#27954)

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

test_shuffle_rows.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

test_top_k_per_row.py

[Bugfix][Kernel] Fix persistent top-k histogram reuse after short rows (#49139)

2026-08-12 12:54:45 -07:00

utils.py

[ROCm][CI] Add More AITER quantization/MoE kernel tests (#49375)

2026-08-05 21:27:33 +00:00