arch(CRITICAL): replace custom qwen3_5.py with base original (1369 lines)

Browse Source

CCCL tuning_rle_encode.cuh AST chain led to reading the base engine zip:
  enginex-vllm-bi100-qwen36-main.zip → qwen3_6_scripts/qwen3_5.py (63KB, 1369 lines)

vs our custom version (85KB, 1780 lines) which added:
  - _hw_policy with hardcoded clamp values
  - nan_to_num(nan=0.0) double disaster
  - Custom _torch_chunk_gated_delta_rule with aggressive clamps
  - Custom FusedMoE fallback logic
  - All of which BROKE the native CoreX acceleration

Sub168 docker log proves:
  - corex_gdn.py:56 loads libcorex_gdn.so (fused GDN decode)
  - corex_gdn.py:228 uses fused GDN prefill
  - corex_moe.py:339 uses CoreX fused MoE (expert-grouped-wmma)
  These are Docker image-internal modules that our custom code never called.

Base original:
  - No nan_to_num (NaN propagates honestly)
  - No custom clamps (uses model weights as-is)
  - Same class structure (Qwen3_5ForCausalLM, Qwen3_5MoeForCausalLM)
  - Docker image's corex modules can intercept through vllm's internal dispatch

qwen3_5_base_original.py kept as reference.

...

This commit is contained in:

Claude

2026-08-08 08:10:55 +00:00

parent 4daa30a267

commit abd3d5640a

2 changed files with 1412 additions and 466 deletions

Show all changes Ignore whitespace when comparing lines Ignore changes in amount of whitespace Ignore changes in whitespace at EOL

Download Patch File Download Diff File Expand all files Collapse all files

1369

qwen3_6_scripts/qwen3_5_base_original.py Normal file

File diff suppressed because it is too large Load Diff