Jingyux/diffusion skip softmax 2

[jingyu-ml]

What does this PR do?

Type of change: ?

Usage

# Add a code snippet demonstrating how to use this

Testing

Before your PR is "Ready for review"

Make sure you read and follow Contributor guidelines and your commits are signed (git commit -s -S).

Make sure you read and follow the Security Best Practices (e.g. avoiding hardcoded trust_remote_code=True, torch.load(..., weights_only=False), pickle, etc.).

• Is this change backward compatible?: ✅ / ❌ / N/A
• If you copied code from any other sources or added a new PIP dependency, did you follow guidance in CONTRIBUTING.md: ✅ / ❌ / N/A
• Did you write any new necessary tests?: ✅ / ❌ / N/A
• Did you update Changelog?: ✅ / ❌ / N/A

Additional Information

[copy-pr-bot]

This pull request requires additional validation before any workflows can run on NVIDIA's runners.

Pull request vetters can view their responsibilities here.

Contributors can view more details about this message here.

[coderabbitai]

Important

Review skipped

Draft detected.

Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro Plus

Run ID: e828fb11-3a08-4d25-b56c-192b11181b31

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

• 🔍 Trigger review

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch jingyux/diffusion-skip-softmax-2

---

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[github-actions]

PR Preview Action v1.8.1
🚀 View preview at https://NVIDIA.github.io/Model-Optimizer/pr-preview/pr-1269/
Built to branch gh-pages at 2026-04-15 22:19 UTC. Preview will be ready when the GitHub Pages deployment is complete.

[codecov]

Codecov Report

❌ Patch coverage is 24.95756% with 442 lines in your changes missing coverage. Please review.
✅ Project coverage is 54.85%. Comparing base (c9b1155) to head (8787d47).
⚠️ Report is 4 commits behind head on main.

Files with missing lines	Patch %	Lines
modelopt/torch/kernels/triton_fa.py	0.00%	108 Missing ⚠️
.../attention_sparsity/methods/triton_skip_softmax.py	16.37%	97 Missing ⚠️
...attention_sparsity/kernels/ltx_triton_attention.py	19.54%	70 Missing ⚠️
...ion_sparsity/kernels/diffusers_triton_attention.py	48.51%	52 Missing ⚠️
...rsity/attention_sparsity/calibration/calibrator.py	9.75%	37 Missing ⚠️
...pt/torch/sparsity/attention_sparsity/conversion.py	40.42%	28 Missing ⚠️
...arsity/attention_sparsity/calibration/calibrate.py	19.04%	17 Missing ⚠️
modelopt/torch/export/unified_export_hf.py	20.00%	16 Missing ⚠️
...y/attention_sparsity/methods/flash_skip_softmax.py	8.33%	11 Missing ⚠️
modelopt/torch/kernels/__init__.py	33.33%	2 Missing ⚠️
... and 4 more

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #1269      +/-   ##
==========================================
- Coverage   60.01%   54.85%   -5.17%     
==========================================
  Files         354      460     +106     
  Lines       40977    49479    +8502     
==========================================
+ Hits        24592    27141    +2549     
- Misses      16385    22338    +5953     

Flag	Coverage Δ
unit	51.76% <24.95%> (-3.95%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ 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.