Revert "[CI] Remove deleted nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-BF16 and its arch aliases"

[mgoin]

Reverts #55026

[claude]

Claude Code Review

This repository is configured for manual code reviews. Comment @claude review for a one-time review, or @claude review always to subscribe this PR to a review on every future push.

_{Tip: disable this comment in your organization's Code Review settings.}

[mgoin]

/ci run

[github-actions]

✅ Triggered Buildkite CI #87303 for commit 11dd797f44ab.

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Repository UI

Review profile: CHILL

Plan: Team

Run ID: c693bb70-8c96-4a75-a0da-9d2dc4af309e

📥 Commits

Reviewing files that changed from the base of the PR and between 5690b02 and 11dd797.

📒 Files selected for processing (3)

• tests/models/registry.py
• vllm/config/speculative.py
• vllm/model_executor/models/registry.py

Included review availability: Your plan provides up to 10 included reviews per hour; 9 remain after this review.

---

📝 Summary

Summary by CodeRabbit

• New Features
  • Added support for three Nemotron-H Omni multimodal reasoning model variants:
    • NemotronH Nano Omni Reasoning V3
    • NemotronH Super Omni Reasoning V3
    • NemotronH Omni Reasoning V3
  • Added support for loading the associated text-model configurations for speculative decoding workflows.
  • Added example configurations for testing and demonstrating these models.

Walkthrough

The change registers three NemotronH Omni reasoning V3 multimodal architectures, adds their example configurations, and makes speculative decoding inspect nested text configurations for two architectures.

Changes

NemotronH Omni reasoning support

Layer / File(s)	Summary
Architecture registration and example configuration vllm/model_executor/models/registry.py, tests/models/registry.py	The multimodal registry maps three NemotronH Omni reasoning architectures to NemotronH_Nano_VL_V2. Example entries add configuration overrides for the Nano variant and mark the Super and standard variants as unavailable online.
Text configuration normalization vllm/config/speculative.py	hf_config_override selects nested text_config for the Super and standard Omni reasoning architectures before MTP detection.

Estimated code review effort: 3 (Moderate) | ~20 minutes

Merge Risk: ⚪ Minimal · up to 11dd7

This change restores support and test examples for three NemotronH Omni reasoning architectures, with nested text configuration handling for speculative decoding where required. No concrete current-head merge risk remains.

Sequence Diagram(s)

sequenceDiagram
  participant ModelRegistry
  participant NemotronH_Nano_VL_V2
  participant hf_config_override
  participant text_config
  ModelRegistry->>NemotronH_Nano_VL_V2: resolve Omni reasoning architecture
  hf_config_override->>text_config: select nested text configuration
  text_config-->>hf_config_override: provide configuration for MTP detection

Loading

Suggested reviewers: zjy0516, khluu, hmellor

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 0.00% which is insufficient. The required threshold is 80.00%. Docstring coverage is scoped to functions touched by this diff. Analyzed 1 functions across 3 files.	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Description check	✅ Passed	The description states that the pull request reverts PR #55026. This matches the restored Nemotron-H Omni reasoning model registrations and architecture aliases in the changeset.
Title check	✅ Passed	The title clearly identifies the pull request as a revert of the change that removed the Nemotron-3-Nano-Omni model and its architecture aliases. This matches the changeset.
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.

• Fix all pre-merge checks with AI

✨ Finishing Touches 💡 1

📝 Generate docstrings 💡

• Create stacked PR
• Commit on current branch

---

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.}