[TRTLLM-13177][doc] Add Nemotron 3 Ultra doc by nv-guomingz · Pull Request #14964 · NVIDIA/TensorRT-LLM

nv-guomingz · 2026-06-04T15:37:00Z

Summary by CodeRabbit

Release Notes

New Features
- Added configuration and deployment support for NVIDIA Nemotron-3-Ultra-550B model with GPU compatibility specifications
Documentation
- Expanded TensorRT-LLM deployment guides to include both Nemotron v3 Super and Ultra model variants
- Updated supported models matrix to reflect Ultra model configuration and feature support

Description

Test Coverage

PR Checklist

Please review the following before submitting your PR:

PR description clearly explains what and why. If using CodeRabbit's summary, please make sure it makes sense.
PR Follows TRT-LLM CODING GUIDELINES to the best of your knowledge.
Test cases are provided for new code paths (see test instructions)
If PR introduces API changes, an appropriate PR label is added - either api-compatible or api-breaking. For api-breaking, include BREAKING in the PR title.
Any new dependencies have been scanned for license and vulnerabilities
CODEOWNERS updated if ownership changes
Documentation updated as needed
Update tava architecture diagram if there is a significant design change in PR.
The reviewers assigned automatically/manually are appropriate for the PR.
Please check this after reviewing the above items as appropriate for this PR.

GitHub Bot Help

To see a list of available CI bot commands, please comment /bot help.

nv-guomingz · 2026-06-04T15:37:28Z

/bot run

coderabbitai · 2026-06-04T15:40:35Z

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: e37d7fab-705b-43cd-b877-96a3eb7ab787

📥 Commits

Reviewing files that changed from the base of the PR and between c17611c and 8a56977.

📒 Files selected for processing (7)

docs/source/_static/config_db.json
docs/source/deployment-guide/deployment-guide-for-nemotron-3-on-trtllm.md
docs/source/deployment-guide/index.rst
docs/source/models/supported-models.md
examples/configs/curated/lookup.yaml
examples/models/core/nemotron/README_nemotron_super_v3.md
scripts/generate_config_table.py

📝 Walkthrough

Walkthrough

This PR expands Nemotron v3 documentation and configuration to support both the Super (120B) and Ultra (550B) model variants on TensorRT-LLM. Configuration database entries, curated lookup mappings, and deployment guide instructions are added or updated for the Ultra model. Documentation indices, feature matrices, and example READMEs are updated to reflect the broader Super & Ultra coverage.

Changes

Nemotron v3 Ultra Documentation and Configuration

Layer / File(s)	Summary
Model configuration registry and metadata `scripts/generate_config_table.py`, `docs/source/_static/config_db.json`, `examples/configs/curated/lookup.yaml`	`MODEL_INFO` script data, `curated_entries` array, and `models` map in config database are populated with Nemotron v3 Ultra model metadata including display name, Hugging Face URL, serve command, GPU compatibility, and throughput scenario. Lookup mapping wires the Ultra model to its curated YAML config.
Deployment guide for Nemotron v3 Super and Ultra `docs/source/deployment-guide/deployment-guide-for-nemotron-3-on-trtllm.md`	Introduction and model description expanded to cover both Super and Ultra variants and their shared hybrid SSM+MoE architecture. Models and GPU requirements sections list separate Super and Ultra checkpoints with expanded Ultra NVFP4 requirements. YAML config selection guidance, server launch commands, and reasoning parser documentation updated for both variants. Example requests, troubleshooting, and benchmarking instructions default to Ultra NVFP4 while covering both models.
Documentation navigation and model feature support updates `docs/source/deployment-guide/index.rst`, `docs/source/models/supported-models.md`, `examples/models/core/nemotron/README_nemotron_super_v3.md`	Deployment guide index toctree updated to reference the expanded Super & Ultra guide. Model feature matrix adds Ultra with enabled disaggregated serving support. Nemotron example README link updated to point to the unified guide URL.

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~10 minutes

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Description check	⚠️ Warning	The PR description is largely a template with no substantive content; critical sections like 'Description' and 'Test Coverage' remain empty or contain only template placeholders.	Fill in the Description section explaining the changes and rationale, and document test coverage details; provide actual implementation notes beyond the template.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Title check	✅ Passed	The PR title clearly summarizes the main change: adding documentation for Nemotron 3 Ultra model to TensorRT-LLM, which aligns with the raw_summary showing doc/config updates for the Ultra variant.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
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.

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

✨ Finishing Touches

🧪 Generate unit tests (beta)

Create PR with unit tests

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

tensorrt-cicd · 2026-06-04T15:43:55Z

PR_Github #52101 [ run ] triggered by Bot. Commit: 8a56977 Link to invocation

tensorrt-cicd · 2026-06-04T15:53:51Z

PR_Github #52101 [ run ] completed with state FAILURE. Commit: 8a56977
/LLM/main/L0_MergeRequest_PR pipeline #41431 completed with status: 'FAILURE'

CI Report

⚠️ Action Required:

Please check the failed tests and fix your PR
If you cannot view the failures, ask the CI triggerer to share details
Once fixed, request an NVIDIA team member to trigger CI again

CI Agent Failure Analysis

Link to invocation

nv-guomingz · 2026-06-05T02:22:49Z

/bot run

tensorrt-cicd · 2026-06-05T02:30:27Z

PR_Github #52208 [ run ] triggered by Bot. Commit: 8a56977 Link to invocation

nv-guomingz · 2026-06-05T04:58:57Z

/bot run

tensorrt-cicd · 2026-06-05T05:05:17Z

PR_Github #52247 [ run ] triggered by Bot. Commit: 2932ae0 Link to invocation

tensorrt-cicd · 2026-06-05T05:06:32Z

PR_Github #52208 [ run ] completed with state ABORTED. Commit: 8a56977

Link to invocation

tensorrt-cicd · 2026-06-05T06:44:22Z

PR_Github #52247 [ run ] completed with state SUCCESS. Commit: 2932ae0
/LLM/main/L0_MergeRequest_PR pipeline #41561 completed with status: 'FAILURE'

CI Report

⚠️ Action Required:

Please check the failed tests and fix your PR
If you cannot view the failures, ask the CI triggerer to share details
Once fixed, request an NVIDIA team member to trigger CI again

CI Agent Failure Analysis

Link to invocation

nv-guomingz · 2026-06-05T07:53:31Z

/bot run

tensorrt-cicd · 2026-06-05T07:59:55Z

PR_Github #52307 [ run ] triggered by Bot. Commit: 2932ae0 Link to invocation

tensorrt-cicd · 2026-06-05T10:58:57Z

PR_Github #52307 [ run ] completed with state SUCCESS. Commit: 2932ae0
/LLM/main/L0_MergeRequest_PR pipeline #41613 completed with status: 'FAILURE'

CI Report

⚠️ Action Required:

Please check the failed tests and fix your PR
If you cannot view the failures, ask the CI triggerer to share details
Once fixed, request an NVIDIA team member to trigger CI again

CI Agent Failure Analysis

Link to invocation

nv-guomingz · 2026-06-05T12:04:02Z

/bot run --disable-fail-fast

tensorrt-cicd · 2026-06-05T12:09:42Z

PR_Github #52337 [ run ] triggered by Bot. Commit: 3bc3218 Link to invocation

tensorrt-cicd · 2026-06-05T17:41:49Z

PR_Github #52337 [ run ] completed with state SUCCESS. Commit: 3bc3218
/LLM/main/L0_MergeRequest_PR pipeline #41642 completed with status: 'FAILURE'

CI Report

⚠️ Action Required:

Please check the failed tests and fix your PR
If you cannot view the failures, ask the CI triggerer to share details
Once fixed, request an NVIDIA team member to trigger CI again

CI Agent Failure Analysis

Link to invocation

Signed-off-by: nv-guomingz <137257613+nv-guomingz@users.noreply.github.com>

nv-guomingz · 2026-06-06T03:17:53Z

/bot run --disable-fail-fast

tensorrt-cicd · 2026-06-06T03:23:59Z

PR_Github #52467 [ run ] triggered by Bot. Commit: e58a676 Link to invocation

tensorrt-cicd · 2026-06-06T07:30:04Z

PR_Github #52467 [ run ] completed with state FAILURE. Commit: e58a676
/LLM/main/L0_MergeRequest_PR pipeline #41759 completed with status: 'FAILURE'

CI Report

⚠️ Action Required:

Please check the failed tests and fix your PR
If you cannot view the failures, ask the CI triggerer to share details
Once fixed, request an NVIDIA team member to trigger CI again

Link to invocation

nv-guomingz · 2026-06-06T13:04:47Z

/bot run --disable-fail-fast

tensorrt-cicd · 2026-06-06T13:11:05Z

PR_Github #52508 [ run ] triggered by Bot. Commit: e58a676 Link to invocation

tensorrt-cicd · 2026-06-06T14:38:03Z

PR_Github #52508 [ run ] completed with state SUCCESS. Commit: e58a676
/LLM/main/L0_MergeRequest_PR pipeline #41799 completed with status: 'FAILURE'

CI Report

⚠️ Action Required:

Please check the failed tests and fix your PR
If you cannot view the failures, ask the CI triggerer to share details
Once fixed, request an NVIDIA team member to trigger CI again

CI Agent Failure Analysis

Link to invocation

nv-guomingz · 2026-06-06T15:55:14Z

/bot run --disable-fail-fast

tensorrt-cicd · 2026-06-06T16:01:11Z

PR_Github #52519 [ run ] triggered by Bot. Commit: e58a676 Link to invocation

tensorrt-cicd · 2026-06-06T16:54:32Z

PR_Github #52519 [ run ] completed with state SUCCESS. Commit: e58a676
/LLM/main/L0_MergeRequest_PR pipeline #41809 completed with status: 'FAILURE'

CI Report

⚠️ Action Required:

Please check the failed tests and fix your PR
If you cannot view the failures, ask the CI triggerer to share details
Once fixed, request an NVIDIA team member to trigger CI again

CI Agent Failure Analysis

Link to invocation

nv-guomingz · 2026-06-07T01:56:26Z

/bot run --disable-fail-fast

tensorrt-cicd · 2026-06-07T02:01:51Z

PR_Github #52538 [ run ] triggered by Bot. Commit: e58a676 Link to invocation

tensorrt-cicd · 2026-06-07T03:11:02Z

PR_Github #52538 [ run ] completed with state SUCCESS. Commit: e58a676
/LLM/main/L0_MergeRequest_PR pipeline #41824 completed with status: 'SUCCESS'

CI Report

Link to invocation

Already updated the code per review comment

Signed-off-by: nv-guomingz <137257613+nv-guomingz@users.noreply.github.com>

nv-guomingz requested review from a team as code owners June 4, 2026 15:37

nv-guomingz requested review from 2ez4bz, laikhtewari and venkywonka June 4, 2026 15:37

github-actions Bot assigned nv-guomingz Jun 4, 2026

2ez4bz approved these changes Jun 4, 2026

View reviewed changes

Comment thread docs/source/deployment-guide/deployment-guide-for-nemotron-3-on-trtllm.md Outdated

Comment thread docs/source/deployment-guide/deployment-guide-for-nemotron-3-on-trtllm.md Outdated

nv-guomingz force-pushed the user/guomingz/update_nemotron_ultra_doc branch 4 times, most recently from 8a2b0f8 to 2932ae0 Compare June 5, 2026 04:58

QiJune approved these changes Jun 5, 2026

View reviewed changes

nv-guomingz force-pushed the user/guomingz/update_nemotron_ultra_doc branch from 2932ae0 to 3bc3218 Compare June 5, 2026 12:03

tijyojwad reviewed Jun 5, 2026

View reviewed changes

Comment thread examples/configs/curated/nemotron-3-ultra-throughput.yaml Outdated

tijyojwad previously requested changes Jun 5, 2026

View reviewed changes

[TRTLLM-13177][doc] Add Nemotron 3 Ultra doc

e58a676

Signed-off-by: nv-guomingz <137257613+nv-guomingz@users.noreply.github.com>

nv-guomingz force-pushed the user/guomingz/update_nemotron_ultra_doc branch from 3bc3218 to e58a676 Compare June 6, 2026 03:17

nv-guomingz merged commit 428cc3e into NVIDIA:main Jun 7, 2026
7 checks passed

This was referenced Jun 8, 2026

[TRTLLM-11548][doc] Add Qwen3.5 deployment guide doc #15111

Merged

[None][doc] Refine Nemotron Ultra doc #15113

Merged

2ez4bz pushed a commit to 2ez4bz/TensorRT-LLM that referenced this pull request Jun 8, 2026

[TRTLLM-13177][doc] Add Nemotron 3 Ultra doc (NVIDIA#14964)

c81dd6f

Signed-off-by: nv-guomingz <137257613+nv-guomingz@users.noreply.github.com>

Conversation

nv-guomingz commented Jun 4, 2026 • edited by coderabbitai Bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary by CodeRabbit

Release Notes

Description

Test Coverage

PR Checklist

GitHub Bot Help

Uh oh!

nv-guomingz commented Jun 4, 2026

Uh oh!

coderabbitai Bot commented Jun 4, 2026

Walkthrough

Changes

Estimated code review effort

❌ Failed checks (1 warning)

Uh oh!

tensorrt-cicd commented Jun 4, 2026

Uh oh!

tensorrt-cicd commented Jun 4, 2026

Uh oh!

Uh oh!

Uh oh!

nv-guomingz commented Jun 5, 2026

Uh oh!

tensorrt-cicd commented Jun 5, 2026

Uh oh!

nv-guomingz commented Jun 5, 2026

Uh oh!

tensorrt-cicd commented Jun 5, 2026

Uh oh!

tensorrt-cicd commented Jun 5, 2026

Uh oh!

tensorrt-cicd commented Jun 5, 2026

Uh oh!

nv-guomingz commented Jun 5, 2026

Uh oh!

tensorrt-cicd commented Jun 5, 2026

Uh oh!

tensorrt-cicd commented Jun 5, 2026

Uh oh!

nv-guomingz commented Jun 5, 2026

Uh oh!

tensorrt-cicd commented Jun 5, 2026

Uh oh!

tensorrt-cicd commented Jun 5, 2026

Uh oh!

Uh oh!

nv-guomingz commented Jun 6, 2026

Uh oh!

tensorrt-cicd commented Jun 6, 2026

Uh oh!

tensorrt-cicd commented Jun 6, 2026

Uh oh!

nv-guomingz commented Jun 6, 2026

Uh oh!

tensorrt-cicd commented Jun 6, 2026

Uh oh!

tensorrt-cicd commented Jun 6, 2026

Uh oh!

nv-guomingz commented Jun 6, 2026

Uh oh!

tensorrt-cicd commented Jun 6, 2026

Uh oh!

tensorrt-cicd commented Jun 6, 2026

Uh oh!

nv-guomingz commented Jun 7, 2026

Uh oh!

tensorrt-cicd commented Jun 7, 2026

Uh oh!

tensorrt-cicd commented Jun 7, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

nv-guomingz commented Jun 4, 2026 •

edited by coderabbitai Bot

Loading