[infra] Add multi-gpu CI for cudax

[pciolkosz]

We could use multi-gpu CI for cudax

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Enterprise

Run ID: 41377cd6-a51f-4746-af69-328b10aa73e5

📥 Commits

Reviewing files that changed from the base of the PR and between 5e49024 and 34a8a8c.

📒 Files selected for processing (1)

• ci/matrix.yaml

---

Note: CodeRabbit is enabled on this repository as a convenience for maintainers
and contributors. Use your best judgment when considering its review comments and
suggestions — a suggested change may be inadequate, unnecessary, or safe to ignore.
Contributors are not expected to address every comment. Human reviews are what
ultimately matter for merging.

Summary

• Added a cudax test job to the pull request CI matrix using the h100_2gpu configuration with sm: 'gpu', enabling multi-GPU CI coverage for cudax.

Walkthrough

Adds one pull_request CI matrix entry for cudax to run test on h100_2gpu with sm set to gpu.

Changes

CI matrix update

Layer / File(s)	Summary
Add cudax h100_2gpu test job ci/matrix.yaml	Adds a pull_request matrix entry for the cudax test job on h100_2gpu with sm: 'gpu'.

Possibly related PRs

• suggestion: NVIDIA/cccl#9435 — also updates ci/matrix.yaml around the h100_2gpu multi-GPU CI matrix entries.

Suggested reviewers

• suggestion: jrhemstad

---

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

[github-actions]

😬 CI Workflow Results

🟥 Finished in 2h 27m: Pass: 99%/508 | Total: 4d 16h | Max: 55m 39s | Hits: 90%/753175

See results here.

[caugonnet]

Thanks so much for setting this @pciolkosz !!