Intermediate checkpointing for sequential calibration

[sugunav14]

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

Summary by CodeRabbit

Release Notes

• New Features
  • Sequential quantization calibration now supports checkpoint save and resume functionality, enabling users to preserve calibration progress and resume from intermediate checkpoints when calibration is interrupted or restarted.
  • New configuration fields control checkpoint directory location and automatic save intervals for flexible checkpointing strategies.
  • HuggingFace models now support seamless checkpoint persistence during sequential calibration.

[copy-pr-bot]

Auto-sync is disabled for draft pull requests in this repository. Workflows must be run manually.

Contributors can view more details about this message here.

[coderabbitai]

Note

Currently processing new changes in this PR. This may take a few minutes, please wait...

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

Run ID: 7693e81d-61c3-4338-9c0f-67d1b730e51b

📥 Commits

Reviewing files that changed from the base of the PR and between ada1e26 and b9035fb.

📒 Files selected for processing (8)

• modelopt/torch/quantization/config.py
• modelopt/torch/quantization/conversion.py
• modelopt/torch/quantization/mode.py
• modelopt/torch/quantization/model_calib.py
• modelopt/torch/quantization/plugins/huggingface.py
• modelopt/torch/quantization/utils/activation_collector.py
• modelopt/torch/quantization/utils/checkpoint.py
• tests/unit/torch/quantization/test_sequential_calibrate.py

 ____________________________________________________________
< To infinity and beyond! Scouring your code for pesky bugs. >
 ------------------------------------------------------------
  \
   \   \
        \ /\
        ( )
      .( o ).

✨ Finishing Touches

📝 Generate docstrings

• Create stacked PR
• Commit on current branch

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch svelury/seq-calib-save-restore

---

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

Tip

You can disable sequence diagrams in the walkthrough.

Disable the reviews.sequence_diagrams setting to disable sequence diagrams in the walkthrough.

[github-actions]

PR Preview Action v1.8.1
🚀 View preview at https://NVIDIA.github.io/Model-Optimizer/pr-preview/pr-1152/
Built to branch gh-pages at 2026-04-01 21:28 UTC. Preview will be ready when the GitHub Pages deployment is complete.

[codecov]

Codecov Report

❌ Patch coverage is 93.84615% with 8 lines in your changes missing coverage. Please review.
✅ Project coverage is 70.31%. Comparing base (ada1e26) to head (0792fab).
⚠️ Report is 5 commits behind head on main.

Files with missing lines	Patch %	Lines
...t/torch/quantization/utils/activation_collector.py	92.42%	5 Missing ⚠️
modelopt/torch/quantization/conversion.py	66.66%	2 Missing ⚠️
modelopt/torch/quantization/model_calib.py	92.85%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #1152      +/-   ##
==========================================
+ Coverage   70.18%   70.31%   +0.13%     
==========================================
  Files         230      231       +1     
  Lines       26080    26213     +133     
==========================================
+ Hits        18304    18432     +128     
- Misses       7776     7781       +5     

☔ 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.