Woosuk Kwon 3b533197a9 [Perf] Parallelize mHC pre-norm JIT warmup (#56683) ...

Signed-off-by: Woosuk Kwon <woosuk@inferact.ai>

2026-09-13 18:58:20 -07:00

..

__init__.py

[Docs] Fix warnings in docs build (#22588)

2025-08-10 05:49:51 -07:00

b12x_warmup.py

[Kernel] Add b12x FP4 MoE backend (#52018)

2026-08-21 08:04:36 -07:00

cutedsl_warmup.py

[2/N][Feat][Perf] Add new warmup infrastructure for JITs. Add predicate filtering for JIT warmup, and migrate Inkling FA4 (#49315)

2026-08-11 07:46:07 -04:00

deep_gemm_warmup.py

[Bugfix] Fix DeepGEMM FP8 warmup coverage (#56452)

2026-09-12 14:20:38 -04:00

flashinfer_autotune_cache.py

[Elastic EP] Reuse CUDA graphs across reconfiguration (#54985)

2026-09-12 10:26:39 -04:00

flashinfer_sparse_mla_warmup.py

[3/N] HiSparse: host-resident sparse-MLA decode hot-buffering (#53781)

2026-09-12 00:36:47 -04:00

jit_warmup_cutedsl_helper.py

[5/N][warmup][DSv4] Migrate NVIDIA CuTeDSL attention kernels (#53566)

2026-09-12 00:05:59 -04:00

jit_warmup_tilelang_helper.py

[9/N][warmup][DSv4] Migrate MHC TileLang kernels (#50178)

2026-09-12 17:25:23 -07:00

jit_warmup_triton_helper.py

[Bugfix] Skip Triton autotune inspection without Triton (#56676)

2026-09-13 17:05:29 +00:00

jit_warmup.py

[Perf] Parallelize mHC pre-norm JIT warmup (#56683)

2026-09-13 18:58:20 -07:00

kernel_warmup.py

[9/N][warmup][DSv4] Migrate MHC TileLang kernels (#50178)

2026-09-12 17:25:23 -07:00

kimi_k3_triton_warmup.py

[New model] Kimi K3 (#50000)

2026-07-30 03:49:58 -07:00

mamba_triton_warmup.py

[Perf] Extend Qwen Triton warmup to avoid first-request latency spikes (#54797)

2026-09-07 11:46:00 +08:00

minimax_m3_msa_warmup.py

[Model] Add MiniMax M3 support (#45381)

2026-06-16 01:01:25 +08:00

qwen4_exp_qsa_warmup.py

[Qwen3.8-Flash-Next] Improve QSA sparse GQA for prefill and short-ctx decode (#54873)

2026-09-04 06:08:02 -07:00

qwen_triton_warmup.py

[Perf] Extend Qwen Triton warmup to avoid first-request latency spikes (#54797)

2026-09-07 11:46:00 +08:00

qwen_vl_triton_warmup.py

[Perf] Extend Qwen Triton warmup to avoid first-request latency spikes (#54797)

2026-09-07 11:46:00 +08:00

replayssm_warmup.py

[Mamba] Add FlashInfer ReplaySSM backend (#52506)

2026-09-01 10:26:43 -04:00

watermark_sample_warmup.py

[watermarking] Dual-key gumbel-max watermarking for speculative decoding support (#56122)

2026-09-12 18:42:01 +00:00