Update Qwen3 235B B300 Configs to match Qwen3 B200 Configs

[rhmukundan]

Summary by CodeRabbit

• Chores
  • Updated configuration settings for model workload optimization.

---

Note: This release contains only internal configuration adjustments with no user-facing changes.

[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]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

Run ID: ebcb689e-d287-4108-aa45-34de7ea40655

📥 Commits

Reviewing files that changed from the base of the PR and between b730ab6 and a759cdf.

📒 Files selected for processing (1)

• scripts/performance/configs/qwen/qwen3_workload_base_configs.py

---

📝 Walkthrough

Walkthrough

This PR modifies QWEN3 workload configurations by disabling moe_a2a_overlap across two configuration variants and removing an explicit virtual_pipeline_model_parallel_size setting from the FP8_CS_V1 configuration.

Changes

Cohort / File(s)	Summary
QWEN3 Workload Configuration scripts/performance/configs/qwen/qwen3_workload_base_configs.py	Disabled moe_a2a_overlap (changed from True to False) in QWEN3_235B_A22B_PRETRAIN_CONFIG_B300_BF16_V1 and QWEN3_235B_A22B_PRETRAIN_CONFIG_B300_FP8_CS_V1 configurations. Removed virtual_pipeline_model_parallel_size=4 from FP8_CS_V1 variant.

Estimated code review effort

🎯 1 (Trivial) | ⏱️ ~3 minutes

Possibly related PRs

• Update Qwen3 30B H100 Base Configs with HybridEP #2477: Modifies the same QWEN3 config file with related changes to moe_a2a_overlap and virtual pipeline parallel settings.
• Update Qwen3 235B A22B MXFP8 GB200/300 recipe and resolve NaN grad norm #2209: Updates the same qwen3 config file with overlapping changes to virtual_pipeline_model_parallel_size and model-parallel behavior.
• cp: Update Qwen3 235B A22B MXFP8 GB200/300 recipe and resolve NaN grad norm (2209) into r0.3.0 #2210: Modifies QWEN3 235B configurations with related adjustments to virtual_pipeline_model_parallel_size and model-parallel settings.

Suggested labels

performance

Suggested reviewers

• malay-nagda
• dingqingy-nv
• ko3n1g

🚥 Pre-merge checks | ✅ 4

✅ Passed checks (4 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title accurately describes the main change: updating Qwen3 235B B300 configurations to match B200 configurations, which aligns with the modifications to moe_a2a_overlap and virtual_pipeline_model_parallel_size settings.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
Test Results For Major Changes	✅ Passed	Configuration-only changes standardizing B300 to match B200 parameters; minor adjustments to parallelism settings without code logic modifications.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

• 📝 Generate docstrings (stacked PR)
• 📝 Generate docstrings (commit on current branch)

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Post copyable unit tests in a comment
• Commit unit tests in branch rmukundan/match_qwen3_235_b300_b200_configs

Tip

Try Coding Plans. Let us write the prompt for your AI agent so you can ship faster (with fewer bugs).
Share your feedback on Discord.

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

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

[yaoyu-33]

explain why in the pr desciption?