[TRTLLM-11318][feat] move VisualGen APIs to a separate dir by zhenhuaw-me · Pull Request #12538 · NVIDIA/TensorRT-LLM

zhenhuaw-me · 2026-03-25T10:27:35Z

Summary by CodeRabbit

Refactor
- Visual Generation API has been reorganized into a dedicated module at tensorrt_llm.visual_gen. Users importing from the top-level tensorrt_llm package are unaffected.
Chores
- Added license headers and updated code ownership assignments.

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)
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.

zhenhuaw-me · 2026-03-25T10:33:08Z

/bot run

coderabbitai · 2026-03-25T10:33:42Z

📝 Walkthrough

Walkthrough

This pull request relocates the VisualGen and VisualGenParams classes from tensorrt_llm/llmapi/visual_gen to a new standalone tensorrt_llm/visual_gen package. All import statements throughout the codebase, top-level exports, documentation, and configuration have been updated to reflect the new module location.

Changes

Cohort / File(s)	Summary
Module Exports `tensorrt_llm/__init__.py`, `tensorrt_llm/llmapi/__init__.py`, `tensorrt_llm/visual_gen/__init__.py`	Removed `VisualGen` and `VisualGenParams` exports from `llmapi`; added new `visual_gen/__init__.py` to expose these classes and `MediaOutput` from the new package location; updated top-level `tensorrt_llm` exports to import from `visual_gen` instead of `llmapi`.
Import Updates `tensorrt_llm/bench/benchmark/visual_gen.py`, `tensorrt_llm/serve/openai_server.py`, `tensorrt_llm/serve/visual_gen_utils.py`, `tests/unittest/_torch/visual_gen/test_visual_gen_args.py`	Updated imports to reference `VisualGen` and `VisualGenParams` from `tensorrt_llm.visual_gen` instead of `tensorrt_llm.llmapi.visual_gen`.
Configuration & Documentation `.github/CODEOWNERS`, `docs/source/models/visual-generation.md`, `tensorrt_llm/visual_gen/visual_gen.py`	Added CODEOWNERS entry assigning `/tensorrt_llm/visual_gen` to the LLM API development team; updated documentation to reflect new module path; added Apache-2.0 license header to `visual_gen.py`.

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~12 minutes

🚥 Pre-merge checks | ✅ 1 | ❌ 2

❌ Failed checks (2 warnings)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 50.00% which is insufficient. The required threshold is 80.00%.	Write docstrings for the functions missing them to satisfy the coverage threshold.
Description check	⚠️ Warning	The PR description is incomplete—it contains only the template with empty placeholder sections and no actual implementation details, rationale, or test coverage information.	Fill in the Description section explaining why VisualGen APIs were moved and the benefits. Provide specific test coverage details and verify all checklist items are addressed.

✅ Passed checks (1 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title clearly and concisely describes the main change: moving VisualGen APIs to a separate directory, matching the actual refactoring across multiple files.

_{✏️ 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.}

coderabbitai

Actionable comments posted: 4

🤖 Prompt for all review comments with AI agents

Verify each finding against the current code and only fix it if needed.

Inline comments:
In `@tensorrt_llm/__init__.py`:
- Line 136: The file header year was not updated when this file was modified;
open the module that exports VisualGen and VisualGenParams (the top of
tensorrt_llm/__init__.py where VisualGen, VisualGenParams are imported) and
update the copyright header year range (replace 2024 with the current year) to
reflect the modification year.

In `@tensorrt_llm/serve/openai_server.py`:
- Line 40: Replace direct class/function imports with module-level imports and
update all usages to reference the module namespace: change "from
tensorrt_llm.llmapi import MultimodalEncoder, tracing" to "import
tensorrt_llm.llmapi as llmapi" and update references to llmapi.MultimodalEncoder
and llmapi.tracing.extract_trace_headers(...); similarly import the visual_gen
module (e.g., "import tensorrt_llm.visual_gen as visual_gen") and replace
VisualGen and VisualGenParams usages with visual_gen.VisualGen and
visual_gen.VisualGenParams so all symbols are accessed through their module
namespaces.

In `@tensorrt_llm/serve/visual_gen_utils.py`:
- Line 12: This file (tensorrt_llm/serve/visual_gen_utils.py) is missing the
required NVIDIA Apache/SPDX header; add the standard NVIDIA Apache 2.0 header
block (including SPDX-License-Identifier: Apache-2.0 and the copyright notice
with the year of the latest meaningful modification) at the top of the file
above the existing imports (e.g., above the line importing VisualGenParams),
ensuring the header format matches other TensorRT-LLM source files.

In `@tensorrt_llm/visual_gen/__init__.py`:
- Line 17: The __all__ export list in __init__.py is not sorted in isort-style
order; update the module-level __all__ to an alphabetically sorted list of the
exported names (VisualGen, VisualGenParams, MediaOutput) so it satisfies Ruff
RUF022 — locate the __all__ declaration in tensorrt_llm.visual_gen.__init__ and
reorder the entries into isort-style alphabetical order.

ℹ️ Review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

Run ID: f025dcb3-6d47-4ad3-850f-cd14ee11a2f6

📥 Commits

Reviewing files that changed from the base of the PR and between 2b5c434 and 28bbad5.

📒 Files selected for processing (10)

.github/CODEOWNERS
docs/source/models/visual-generation.md
tensorrt_llm/__init__.py
tensorrt_llm/bench/benchmark/visual_gen.py
tensorrt_llm/llmapi/__init__.py
tensorrt_llm/serve/openai_server.py
tensorrt_llm/serve/visual_gen_utils.py
tensorrt_llm/visual_gen/__init__.py
tensorrt_llm/visual_gen/visual_gen.py
tests/unittest/_torch/visual_gen/test_visual_gen_args.py

💤 Files with no reviewable changes (1)

tensorrt_llm/llmapi/init.py

tensorrt-cicd · 2026-03-25T10:53:34Z

PR_Github #40319 [ run ] triggered by Bot. Commit: 28bbad5 Link to invocation

tensorrt-cicd · 2026-03-25T20:41:47Z

PR_Github #40319 [ run ] completed with state SUCCESS. Commit: 28bbad5
/LLM/main/L0_MergeRequest_PR pipeline #31429 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

QiJune

LGTM

zhenhuaw-me · 2026-03-31T13:52:17Z

/bot run

tensorrt-cicd · 2026-03-31T13:57:56Z

PR_Github #40948 [ run ] triggered by Bot. Commit: 214c517 Link to invocation

tensorrt-cicd · 2026-03-31T15:34:41Z

PR_Github #40948 [ run ] completed with state FAILURE. Commit: 214c517
/LLM/main/L0_MergeRequest_PR pipeline #31936 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

tensorrt-cicd · 2026-04-01T00:50:19Z

PR_Github #41043 [ run ] completed with state FAILURE. Commit: fe79590
/LLM/main/L0_MergeRequest_PR pipeline #32022 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

zhenhuaw-me · 2026-04-01T02:27:43Z

/bot run

tensorrt-cicd · 2026-04-01T02:33:36Z

PR_Github #41078 [ run ] triggered by Bot. Commit: 7c75288 Link to invocation

tensorrt-cicd · 2026-04-01T04:45:23Z

PR_Github #41078 [ run ] completed with state FAILURE. Commit: 7c75288
/LLM/main/L0_MergeRequest_PR pipeline #32054 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

zhenhuaw-me · 2026-04-01T06:17:26Z

/bot run --disable-fail-fast

tensorrt-cicd · 2026-04-01T06:22:56Z

PR_Github #41129 [ run ] triggered by Bot. Commit: d10520b Link to invocation

tensorrt-cicd · 2026-04-01T14:18:42Z

PR_Github #41129 [ run ] completed with state SUCCESS. Commit: d10520b
/LLM/main/L0_MergeRequest_PR pipeline #32100 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

zhenhuaw-me · 2026-04-02T02:51:06Z

/bot run --disable-fail-fast

tensorrt-cicd · 2026-04-02T02:57:16Z

PR_Github #41315 [ run ] triggered by Bot. Commit: 21f6e38 Link to invocation

zhenhuaw-me · 2026-04-02T10:13:46Z

/bot run --disable-fail-fast

tensorrt-cicd · 2026-04-02T10:25:45Z

PR_Github #41416 [ run ] triggered by Bot. Commit: dd55c23 Link to invocation

Signed-off-by: Zhenhua Wang <[email protected]>

zhenhuaw-me · 2026-04-02T15:42:28Z

/bot run --disable-fail-fast

tensorrt-cicd · 2026-04-02T15:47:55Z

PR_Github #41456 [ run ] triggered by Bot. Commit: 5b0f192 Link to invocation

tensorrt-cicd · 2026-04-02T20:38:06Z

PR_Github #41456 [ run ] completed with state SUCCESS. Commit: 5b0f192
/LLM/main/L0_MergeRequest_PR pipeline #32387 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

zhenhuaw-me · 2026-04-03T01:33:47Z

/bot run --disable-fail-fast

tensorrt-cicd · 2026-04-03T01:40:26Z

PR_Github #41537 [ run ] triggered by Bot. Commit: 5b0f192 Link to invocation

zhenhuaw-me · 2026-04-03T01:42:02Z

/bot skip --comment "Failing autodeploy DS and TensorRT backend test is unrelated to this PRs changes"

tensorrt-cicd · 2026-04-03T01:47:55Z

PR_Github #41540 [ skip ] triggered by Bot. Commit: 5b0f192 Link to invocation

tensorrt-cicd · 2026-04-03T01:54:45Z

PR_Github #41540 [ skip ] completed with state SUCCESS. Commit: 5b0f192
Skipping testing for commit 5b0f192

Link to invocation

) Signed-off-by: Zhenhua Wang <[email protected]>

zhenhuaw-me requested review from a team as code owners March 25, 2026 10:27

zhenhuaw-me requested review from FrankD412, hchings, kaiyux, laikhtewari, tburt-nv and yuanjingx87 March 25, 2026 10:27

github-actions Bot assigned zhenhuaw-me Mar 25, 2026

zhenhuaw-me requested review from QiJune, Superjomn, chang-l, litaotju and schetlur-nv March 25, 2026 10:29

zhenhuaw-me added the VisualGen label Mar 25, 2026

coderabbitai Bot reviewed Mar 25, 2026

View reviewed changes

Comment thread tensorrt_llm/__init__.py

Comment thread tensorrt_llm/serve/openai_server.py

Comment thread tensorrt_llm/serve/visual_gen_utils.py

Comment thread tensorrt_llm/visual_gen/__init__.py

QiJune approved these changes Mar 31, 2026

View reviewed changes

zhenhuaw-me force-pushed the move-visualgen-apis branch from 28bbad5 to 214c517 Compare March 31, 2026 13:51

zhenhuaw-me force-pushed the move-visualgen-apis branch from fe79590 to 7c75288 Compare April 1, 2026 02:27

ZhanruiSunCh approved these changes Apr 1, 2026

View reviewed changes

zhenhuaw-me enabled auto-merge (squash) April 1, 2026 06:16

zhenhuaw-me force-pushed the move-visualgen-apis branch from d10520b to 21f6e38 Compare April 2, 2026 02:50

zhenhuaw-me force-pushed the move-visualgen-apis branch from 21f6e38 to dd55c23 Compare April 2, 2026 10:13

[TRTLLM-11318][feat] move VisualGen APIs to a separate dir

5b0f192

Signed-off-by: Zhenhua Wang <[email protected]>

zhenhuaw-me force-pushed the move-visualgen-apis branch from dd55c23 to 5b0f192 Compare April 2, 2026 15:41

zhenhuaw-me merged commit 31fbf0f into NVIDIA:main Apr 3, 2026
5 checks passed

zhenhuaw-me deleted the move-visualgen-apis branch April 3, 2026 01:54

govind-ramnarayan pushed a commit to nv-auto-deploy/TensorRT-LLM that referenced this pull request Apr 6, 2026

[TRTLLM-11318][feat] move VisualGen APIs to a separate dir (NVIDIA#12538

d519adc

) Signed-off-by: Zhenhua Wang <[email protected]>

karen-sy pushed a commit to karen-sy/TensorRT-LLM that referenced this pull request Apr 7, 2026

[TRTLLM-11318][feat] move VisualGen APIs to a separate dir (NVIDIA#12538

c4b1381

) Signed-off-by: Zhenhua Wang <[email protected]>

Conversation

zhenhuaw-me commented Mar 25, 2026 • edited by coderabbitai Bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary by CodeRabbit

Description

Test Coverage

PR Checklist

GitHub Bot Help

Uh oh!

zhenhuaw-me commented Mar 25, 2026

Uh oh!

coderabbitai Bot commented Mar 25, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Walkthrough

Changes

Estimated code review effort

❌ Failed checks (2 warnings)

Uh oh!

coderabbitai Bot left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

tensorrt-cicd commented Mar 25, 2026

Uh oh!

tensorrt-cicd commented Mar 25, 2026

Uh oh!

QiJune left a comment

Choose a reason for hiding this comment

Uh oh!

zhenhuaw-me commented Mar 31, 2026

Uh oh!

tensorrt-cicd commented Mar 31, 2026

Uh oh!

tensorrt-cicd commented Mar 31, 2026

Uh oh!

tensorrt-cicd commented Apr 1, 2026

Uh oh!

zhenhuaw-me commented Apr 1, 2026

Uh oh!

tensorrt-cicd commented Apr 1, 2026

Uh oh!

tensorrt-cicd commented Apr 1, 2026

Uh oh!

zhenhuaw-me commented Apr 1, 2026

Uh oh!

tensorrt-cicd commented Apr 1, 2026

Uh oh!

tensorrt-cicd commented Apr 1, 2026

Uh oh!

zhenhuaw-me commented Apr 2, 2026

Uh oh!

tensorrt-cicd commented Apr 2, 2026

Uh oh!

zhenhuaw-me commented Apr 2, 2026

Uh oh!

tensorrt-cicd commented Apr 2, 2026

Uh oh!

zhenhuaw-me commented Apr 2, 2026

Uh oh!

tensorrt-cicd commented Apr 2, 2026

Uh oh!

tensorrt-cicd commented Apr 2, 2026

Uh oh!

zhenhuaw-me commented Apr 3, 2026

Uh oh!

tensorrt-cicd commented Apr 3, 2026

Uh oh!

zhenhuaw-me commented Apr 3, 2026

Uh oh!

tensorrt-cicd commented Apr 3, 2026

Uh oh!

tensorrt-cicd commented Apr 3, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

zhenhuaw-me commented Mar 25, 2026 •

edited by coderabbitai Bot

Loading

coderabbitai Bot commented Mar 25, 2026 •

edited

Loading