docs(qwen3_8): update NVFP4 checkpoint reference to nvidia/Qwen3.8-27B-NVFP4

[zhenshanx-nv]

Background

families/qwen3_8's NVFP4 code comments and docstrings referenced RadixArk/Qwen3.8-27B-NVFP4 as the example checkpoint. NVIDIA has since published its own official nvidia/Qwen3.8-27B-NVFP4 checkpoint -- the same ModelOpt MIXED_PRECISION export, verified byte-identical in structure and behavior (weight loading, build time, engine size, runtime, accuracy) -- and that is now the checkpoint tested against and recommended.

Exit Criteria

• No functional change: calibrate_qwen3_8_nvfp4() never hardcoded either checkpoint name (it reads generic HF tensor names from whatever model_dir is passed in), so this PR is comments/docstrings only.
• The three references to RadixArk/Qwen3.8-27B-NVFP4 in quantization.py/checkpoint_mapper.py now name nvidia/Qwen3.8-27B-NVFP4 instead.

Implementation

Straightforward string replacement in a module docstring, an inline comment, and a runtime error message.

Change categories

• Documentation only
• Model or runtime behavior
• Public API
• ABI
• Bundle or artifact format
• Dependencies
• CI or developer tooling

Validation

Commands and Results

• ruff check --config ruff.toml families/qwen3_8/quantization.py families/qwen3_8/checkpoint_mapper.py -- all checks passed.
• git diff reviewed: exactly 3 lines changed, no code logic touched.

Hardware, Environment, and Revisions

• Repository head: this branch, based on upstream/main.
• Not applicable: no runtime behavior change, no build/GPU testing needed.

Not Run / Remaining Gaps

• None: this is a comments/docstrings-only change with no code path affected.

Contributor Self-Review

• I have completed a self-review of this change.

Notes For Future Readers

None.

Risk level

• Low
• Medium
• High

Comments and docstrings only; no code logic changed.

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Repository: NVIDIA/TensorRT-Model-Connect/.coderabbit.yaml

Review profile: CHILL

Plan: Enterprise

Run ID: e794c64d-2f67-491d-909f-71cea850c90a

📥 Commits

Reviewing files that changed from the base of the PR and between 393ab02 and 28fef18.

📒 Files selected for processing (2)

• families/qwen3_8/checkpoint_mapper.py
• families/qwen3_8/quantization.py

Included review availability: Your plan provides up to 12 included reviews per hour; 11 remain after this review.

---

📝 Summary

Summary

Updates Qwen3.8 documentation and error-message references to NVIDIA’s official nvidia/Qwen3.8-27B-NVFP4 checkpoint. The change affects comments, module documentation, and one diagnostic string only. No runtime or quantization behavior changes.

Architecture impact

• Family-owned files: families/qwen3_8/checkpoint_mapper.py and families/qwen3_8/quantization.py.
• Shared surfaces: None identified.
• Dependency directions: No new dependencies or cross-family dependencies.
• Affected consumers: Qwen3.8 checkpoint mapping and NVFP4 quantization diagnostics.
• Unresolved blast-radius questions: None identified from the reviewed changes.

Review

PASS — No standards or specification violation was found in the reviewed evidence.

Review finding counts are unavailable because no current review findings were supplied. Test results were not supplied.

Walkthrough

Changes

Qwen3.8 checkpoint naming

Layer / File(s)	Summary
Update checkpoint references families/qwen3_8/checkpoint_mapper.py, families/qwen3_8/quantization.py	Documentation and the missing-quantized-tensor error message now reference nvidia/Qwen3.8-27B-NVFP4. Quantization behavior is unchanged.

Priority: ⬇️ Low

Estimated code review effort: 1 (Trivial) | ~3 minutes

Change: Other

🚥 Pre-merge checks | ✅ 9

✅ Passed checks (9 passed)

Check name	Status	Explanation
Docstring Coverage	✅ Passed	Docstring coverage is 100.00% which is sufficient. The required threshold is 80.00%. Docstring coverage is scoped to functions touched by this diff. Analyzed 1 functions across 2 files.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Family Ownership Boundary	✅ Passed	PASS. The pull request changes only three references within families/qwen3_8: the comment at checkpoint_mapper.py:169, the module docstring at quantization.py:6, and the error text at `quantizat…
Shared Semantic Neutrality	✅ Passed	PASS. The review range changes only families/qwen3_8/checkpoint_mapper.py and families/qwen3_8/quantization.py. These are family-owned Python files, which the check excludes. The three changed lin…
Benchmark Validation Integrity	✅ Passed	PASS. The review-scoped diff changes only three checkpoint-name strings in two Python files: one comment, one module docstring, and one runtime error message. The calibration function and its tensor-r…
Shared Change Blast Radius	✅ Passed	PASS. The pull request changes only two files under families/qwen3_8. The exact diff contains three checkpoint-name replacements in a comment, a module docstring, and an error-message string. No sha…
Description check	✅ Passed	The description covers the background, exit criteria, implementation, change category, validation, environment, remaining gaps, self-review, notes, and risk level. It clearly states that the change up…
Title check	✅ Passed	The title clearly and concisely identifies the Qwen3.8 documentation change and the updated NVIDIA NVFP4 checkpoint reference.

---

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