[None][fix] Add missing absolute pe in Qwen3-VL Vision Encoder

[Nekofish-L]

Description
This PR fixes a bug in the Qwen3VL encoder forward pass where pos_embeds were not being added to the hidden_states after the embedding layer.

Why Unit Tests Failed to Catch This: The current unittests use a randomly initialized model. The pos_embed weights are initialized to zeros or very small values. As a result, adding them doesn't produce a detectable difference in output tensors.
I conducted tests using the full Qwen3-VL-8B-Instruct pre-trained weights. By comparing the output of this implementation against the reference HF transformers output, the results are as follows:

• before fix (diff):

	max abs diff	mean abs diff
Context phase	3.625	0.675
Generation phase	1.125	0.135

• after fix (diff):

	max abs diff	mean abs diff
Context phase	0.359	0.058
Generation phase	0.25	0.032

Summary by CodeRabbit

• New Features
  • Enhanced positional embedding computation by introducing interpolated grid-based embeddings alongside existing rotary embeddings, providing more comprehensive spatial conditioning for improved model accuracy.

_{✏️ Tip: You can customize this high-level summary in your review settings.}

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

/bot [-h] ['run', 'kill', 'skip', 'reuse-pipeline'] ...

Provide a user friendly way for developers to interact with a Jenkins server.

Run /bot [-h|--help] to print this help message.

See details below for each supported subcommand.

Details

run [--reuse-test (optional)pipeline-id --disable-fail-fast --skip-test --stage-list "A10-PyTorch-1, xxx" --gpu-type "A30, H100_PCIe" --test-backend "pytorch, cpp" --add-multi-gpu-test --only-multi-gpu-test --disable-multi-gpu-test --post-merge --extra-stage "H100_PCIe-TensorRT-Post-Merge-1, xxx" --detailed-log --debug(experimental)]

Launch build/test pipelines. All previously running jobs will be killed.

--reuse-test (optional)pipeline-id (OPTIONAL) : Allow the new pipeline to reuse build artifacts and skip successful test stages from a specified pipeline or the last pipeline if no pipeline-id is indicated. If the Git commit ID has changed, this option will be always ignored. The DEFAULT behavior of the bot is to reuse build artifacts and successful test results from the last pipeline.

--disable-reuse-test (OPTIONAL) : Explicitly prevent the pipeline from reusing build artifacts and skipping successful test stages from a previous pipeline. Ensure that all builds and tests are run regardless of previous successes.

--disable-fail-fast (OPTIONAL) : Disable fail fast on build/tests/infra failures.

--skip-test (OPTIONAL) : Skip all test stages, but still run build stages, package stages and sanity check stages. Note: Does NOT update GitHub check status.

--stage-list "A10-PyTorch-1, xxx" (OPTIONAL) : Only run the specified test stages. Examples: "A10-PyTorch-1, xxx". Note: Does NOT update GitHub check status.

--gpu-type "A30, H100_PCIe" (OPTIONAL) : Only run the test stages on the specified GPU types. Examples: "A30, H100_PCIe". Note: Does NOT update GitHub check status.

--test-backend "pytorch, cpp" (OPTIONAL) : Skip test stages which don't match the specified backends. Only support [pytorch, cpp, tensorrt, triton]. Examples: "pytorch, cpp" (does not run test stages with tensorrt or triton backend). Note: Does NOT update GitHub pipeline status.

--only-multi-gpu-test (OPTIONAL) : Only run the multi-GPU tests. Note: Does NOT update GitHub check status.

--disable-multi-gpu-test (OPTIONAL) : Disable the multi-GPU tests. Note: Does NOT update GitHub check status.

--add-multi-gpu-test (OPTIONAL) : Force run the multi-GPU tests in addition to running L0 pre-merge pipeline.

--post-merge (OPTIONAL) : Run the L0 post-merge pipeline instead of the ordinary L0 pre-merge pipeline.

--extra-stage "H100_PCIe-TensorRT-Post-Merge-1, xxx" (OPTIONAL) : Run the ordinary L0 pre-merge pipeline and specified test stages. Examples: --extra-stage "H100_PCIe-TensorRT-Post-Merge-1, xxx".

--detailed-log (OPTIONAL) : Enable flushing out all logs to the Jenkins console. This will significantly increase the log volume and may slow down the job.

--debug (OPTIONAL) : Experimental feature. Enable access to the CI container for debugging purpose. Note: Specify exactly one stage in the stage-list parameter to access the appropriate container environment. Note: Does NOT update GitHub check status.

For guidance on mapping tests to stage names, see docs/source/reference/ci-overview.md
and the scripts/test_to_stage_mapping.py helper.

kill

kill

Kill all running builds associated with pull request.

skip

skip --comment COMMENT

Skip testing for latest commit on pull request. --comment "Reason for skipping build/test" is required. IMPORTANT NOTE: This is dangerous since lack of user care and validation can cause top of tree to break.

reuse-pipeline

reuse-pipeline

Reuse a previous pipeline to validate current commit. This action will also kill all currently running builds associated with the pull request. IMPORTANT NOTE: This is dangerous since lack of user care and validation can cause top of tree to break.

[coderabbitai]

📝 Walkthrough

Walkthrough

A new positional embedding computation path using fast_pos_embed_interpolate(grid_thw) is added to the forward method. The computed positional embeddings are combined with patch-based hidden states before subsequent processing, complementing the existing rotary positional embeddings.

Changes

Cohort / File(s)	Summary
Qwen3VL Model Updates tensorrt_llm/_torch/models/modeling_qwen3vl.py	Adds interpolated grid-based positional embeddings to hidden states in forward method via fast_pos_embed_interpolate(grid_thw) and combines them with patch embeddings

Estimated code review effort

🎯 3 (Moderate) | ⏱️ ~20 minutes

🚥 Pre-merge checks | ✅ 2 | ❌ 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%.	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (2 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title clearly specifies the fix being applied: adding missing absolute positional embeddings in the Qwen3-VL Vision Encoder, directly matching the changeset.
Description check	✅ Passed	The PR description clearly explains the bug, provides empirical evidence of the fix's effectiveness, and explains why tests failed to catch it.

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

✨ Finishing touches

• 📝 Generate docstrings

---

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.}

[Nekofish-L]

Hi @yechank-nvidia , could you please take a look at this PR?

[yechank-nvidia]

Nice catch! Thanks for the PR!

[yechank-nvidia]

/bot run

[tensorrt-cicd]

PR_Github #33882 [ run ] triggered by Bot. Commit: 25db9c0

[tensorrt-cicd]

PR_Github #33882 [ run ] completed with state SUCCESS. Commit: 25db9c0
/LLM/main/L0_MergeRequest_PR pipeline #26128 completed with status: 'FAILURE'

⚠️ 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

[yechank-nvidia]

/bot run

[tensorrt-cicd]

PR_Github #33954 [ run ] triggered by Bot. Commit: 0aea0d7

[tensorrt-cicd]

PR_Github #33954 [ run ] completed with state SUCCESS. Commit: 0aea0d7
/LLM/main/L0_MergeRequest_PR pipeline #26190 completed with status: 'FAILURE'

⚠️ 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

[yechank-nvidia]

/bot run

[tensorrt-cicd]

PR_Github #33985 [ run ] triggered by Bot. Commit: 0aea0d7

[tensorrt-cicd]

PR_Github #33985 [ run ] completed with state SUCCESS. Commit: 0aea0d7
/LLM/main/L0_MergeRequest_PR pipeline #26217 completed with status: 'FAILURE'

⚠️ 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

[yechank-nvidia]

/bot run

[tensorrt-cicd]

PR_Github #34137 [ run ] triggered by Bot. Commit: 0aea0d7

[tensorrt-cicd]

PR_Github #34137 [ run ] completed with state SUCCESS. Commit: 0aea0d7
/LLM/main/L0_MergeRequest_PR pipeline #26337 completed with status: 'SUCCESS'

[jaedeok-nvidia]

Thanks @Nekofish-L for the fix 👍

@yechank-nvidia This is a fix of qwen3-vl functionality. Would you please help cherry-pick this PR into release/1.2?

cc. @2ez4bz @chang-l