Skip to content

Conversation

@gramalingam
Copy link
Collaborator

@gramalingam gramalingam commented Oct 16, 2025

Gemma3 applies an extra (simplified) normalization to query and key before the rotary embedding. Extend GQA fusion to handle this.

TODO: will add test-case separately.

Signed-off-by: Ganesan Ramalingam <[email protected]>
@codecov
Copy link

codecov bot commented Oct 16, 2025

Codecov Report

❌ Patch coverage is 22.22222% with 14 lines in your changes missing coverage. Please review.
✅ Project coverage is 70.35%. Comparing base (8089bc7) to head (7c5f947).
⚠️ Report is 1 commits behind head on main.
✅ All tests successful. No failed tests found.

Files with missing lines Patch % Lines
onnxscript/rewriter/ort_fusions/gqa.py 22.22% 12 Missing and 2 partials ⚠️
Additional details and impacted files
@@            Coverage Diff             @@
##             main    #2639      +/-   ##
==========================================
- Coverage   70.39%   70.35%   -0.04%     
==========================================
  Files         222      222              
  Lines       26294    26312      +18     
  Branches     2630     2632       +2     
==========================================
+ Hits        18509    18513       +4     
- Misses       6865     6877      +12     
- Partials      920      922       +2     

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

@gramalingam gramalingam marked this pull request as ready for review October 16, 2025 21:33
@gramalingam gramalingam merged commit dd8cb69 into main Oct 16, 2025
32 checks passed
@gramalingam gramalingam deleted the rama/gqafix branch October 16, 2025 21:56
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

Development

Successfully merging this pull request may close these issues.

2 participants