Skip to content

Refactor Qwen3.5 MoE quantization to use _QuantFunctionalMixin#1170

Closed
cjluo-nv wants to merge 3 commits into
mainfrom
chenjiel/refactor_qwen35
Closed

Refactor Qwen3.5 MoE quantization to use _QuantFunctionalMixin#1170
cjluo-nv wants to merge 3 commits into
mainfrom
chenjiel/refactor_qwen35

Reviewer 2

59d10d9
Select commit
Loading
Failed to load commit list.
Codecov / codecov/project succeeded Apr 2, 2026 in 0s

75.74% (+1.47%) compared to 00c002f

View this Pull Request on Codecov

75.74% (+1.47%) compared to 00c002f

Details

Codecov Report

❌ Patch coverage is 11.70213% with 83 lines in your changes missing coverage. Please review.
✅ Project coverage is 75.74%. Comparing base (00c002f) to head (59d10d9).
⚠️ Report is 3 commits behind head on main.

Files with missing lines Patch % Lines
modelopt/torch/export/moe_utils.py 3.70% 52 Missing ⚠️
modelopt/torch/quantization/plugins/huggingface.py 18.18% 27 Missing ⚠️
modelopt/torch/export/unified_export_hf.py 33.33% 4 Missing ⚠️
Additional details and impacted files 
@@            Coverage Diff             @@
##             main    #1170      +/-   ##
==========================================
+ Coverage   74.27%   75.74%   +1.47%     
==========================================
  Files         349      349              
  Lines       39846    39886      +40     
==========================================
+ Hits        29594    30212     +618     
+ Misses      10252     9674     -578

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:
  • ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
View more details on Codecov