Claude a8ca42b59c perf: hgemm_warptiling Config B — beats cublas on MoE-sized GEMM (0.7x) ...

probe_k10_configs.sh results on BI-V100:
  256x4096 @ 4096x11008:
    cublas:   10.554 ms
    Config B:  7.649 ms (0.7x cublas — FASTER)
    Config A:  2308 ms  (old broken config)

Config B: BM128 BN128 BK16 WM64 WN64 WNITER2 TM8 TN4 NT128
Root cause of Config A slowness: WN=128 WNITER=4 caused
excessive register pressure and smem bank conflicts.

2026-08-15 05:11:32 +00:00

..

csrc

feat: moe_expert_gemm.cpp — C++ loop over experts via ixformer_linear (replaces Python for-loop)

2026-08-14 12:09:50 +00:00

fla_kernels

feat: import CUDA kernels from xllm/CCCL/FLA upstream repos

2026-08-14 07:48:52 +00:00

include

fix(CRITICAL): engine death on image request + stop overwriting base corex modules

2026-08-10 09:12:05 +00:00

python

fix(build): match project_7 proven Dockerfile — remove ENV lines + .dockerignore

2026-08-13 02:48:28 +00:00

xllm_kernels

perf: hgemm_warptiling Config B — beats cublas on MoE-sized GEMM (0.7x)

2026-08-15 05:11:32 +00:00

xllm_layers

fix(critical): fold max_completion_tokens + max_num_seqs=2 + max_model_len=80000 + xllm_latest layer import

2026-08-13 03:19:39 +00:00

xllm_models

fix(critical): fold max_completion_tokens + max_num_seqs=2 + max_model_len=80000 + xllm_latest layer import

2026-08-13 03:19:39 +00:00

build.sh

fix(build): 回退qwen3_6_scripts+ex_engine到26e6cb40(能得分版本)

2026-08-12 01:33:24 +00:00

deploy_corex_modules.sh

feat(EX): corex_fa2.py — third dlopen module from comp 168 AST chain

2026-08-10 04:01:21 +00:00

precompile_moe_kernels.py

fix(build): 回退qwen3_6_scripts+ex_engine到26e6cb40(能得分版本)

2026-08-12 01:33:24 +00:00

precompile_moe_topk.py

fix(MoE): robust CUDA kernel loading + no-GPU precompile

2026-08-10 07:50:28 +00:00