[TRTLLM-12982][perf] reuse multi-item scoring position_ids and params by ixlmar · Pull Request #15413 · NVIDIA/TensorRT-LLM

ixlmar · 2026-06-16T10:24:42Z

Description

Implements performance improvements for multi-item scoring:

FlashInferMultiItemParams are reused across model layers
position_ids are computed on GPU and reused across model layers

Note for reviewers: Changeset contains moved code, so consider something like git diff --color-moved=dimmed-zebra -w.

Test Coverage

Covered by existing tests.

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.

Summary by CodeRabbit

Refactor
- Simplified the attention module API by removing multi_item_part_lens parameter from the public interface.
- Relocated multi-item inference configuration to metadata preparation stage for better separation of concerns.
- Consolidated multi-item scoring validation to the metadata layer across attention backends.
Performance
- Added NVTX instrumentation for improved profiling of attention mechanisms.
Improvements
- Enhanced encoder-only inference with improved multi-item batching support.

ixlmar · 2026-06-16T10:24:55Z

/bot run --disable-fail-fast

coderabbitai · 2026-06-16T10:29:00Z

📝 Walkthrough

Walkthrough

multi_item_part_lens is moved from Attention.forward/forward_impl/AttentionForwardArgs into AttentionMetadata.prepare() as a keyword-only argument. FlashInfer metadata gains _process_multi_item_part_lens and stores results in _multi_item_params; other backends reject the argument at prepare time. encode() in llm.py now computes position_ids from multi_item_part_lens, and NVTX profiling wraps the plan call.

Changes

Multi-item part_lens: prepare() ownership refactor

Layer / File(s)	Summary
Base interface contract: `prepare()` signature and `AttentionForwardArgs` cleanup `tensorrt_llm/_torch/attention_backend/interface.py`	`AttentionMetadata.prepare` gains a keyword-only `multi_item_part_lens` parameter; the `multi_item_part_lens` field is removed from `AttentionForwardArgs`.
FlashInfer metadata: `_process_multi_item_part_lens`, `prepare()`, `plan()`, NVTX `tensorrt_llm/_torch/attention_backend/flashinfer.py`	New `_multi_item_params` field and `_process_multi_item_part_lens` helper build GPU masking tensors; `prepare()` stores results in context-only path; `plan()` drops its `multi_item_params` arg and reads `self._multi_item_params`; `forward_impl` drops the parameter; `metadata.plan()` call is wrapped in `nvtx_range`.
Other backends: `prepare()` validation `tensorrt_llm/_torch/attention_backend/star_flashinfer.py`, `tensorrt_llm/_torch/attention_backend/trtllm.py`, `tensorrt_llm/_torch/attention_backend/vanilla.py`	`StarAttentionMetadata`, `TrtllmAttentionMetadata`, `VanillaAttentionMetadata` each add the `multi_item_part_lens` keyword arg to `prepare()` and raise `ValueError` when non-None; `TrtllmAttentionMetadata.prepare_encoder_only` gets the same; forward-time validation guards are removed from all three backends.
`Attention` module: remove `multi_item_part_lens` from public API and RoPE logic `tensorrt_llm/_torch/modules/attention.py`	`Attention._attn_impl`, `forward_impl`, and `forward` lose the `multi_item_part_lens` parameter; both `AttentionForwardArgs` construction sites drop the argument; the RoPE `position_ids` rewrite block keyed on `multi_item_part_lens` is deleted.
Encode path wiring: `position_ids` construction and model engine forwarding `tensorrt_llm/_torch/pyexecutor/model_engine.py`, `tensorrt_llm/llmapi/llm.py`	`model_engine.py` reads `multi_item_part_lens` from inputs and forwards it into `prepare_encoder_only`/`prepare`; CUDA graph mode asserts it is `None`; `llm.py`'s `encode()` adds `@torch.inference_mode()`, imports `torch_multi_arange`/`prefer_pinned`, and constructs packed `position_ids` from `multi_item_part_lens` before `batch_forward`.

Estimated code review effort

🎯 4 (Complex) | ⏱️ ~45 minutes

Possibly related PRs

NVIDIA/TensorRT-LLM#14693: Directly related — both PRs modify flashinfer.py to implement multi-item scoring support by reshaping how multi_item_part_lens/multi_item_params are created, stored, and consumed across plan()/forward_impl.

Suggested reviewers

tburt-nv
Funatiq
brb-nv

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Title check	✅ Passed	The PR title clearly and specifically describes the main changes: reusing multi-item scoring position_ids and params for performance improvement, with proper JIRA ticket and performance type indicator.
Description check	✅ Passed	The PR description adequately explains what is being changed (reuse of FlashInferMultiItemParams and position_ids) and why (performance improvement), with test coverage and a completed checklist. All critical sections are present.
Docstring Coverage	✅ Passed	Docstring coverage is 100.00% which is sufficient. The required threshold is 80.00%.
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.}

coderabbitai

Actionable comments posted: 1

🧹 Nitpick comments (1)

tensorrt_llm/_torch/attention_backend/flashinfer.py (1)

766-770: 💤 Low value

Consider adding strict=True to zip() and using iterable unpacking.

Static analysis flagged two minor style issues:

Line 766-767: zip() without strict= parameter. While both iterables are guaranteed same-length by construction, adding strict=True improves defensiveness.
Line 769-770: List concatenation can use iterable unpacking for clarity.

♻️ Optional style improvements

         range_ends = torch.tensor(
             [
                 item_len + 1
                 for req_part_lens, token_pos_in_items_raw_len in zip(
-                    multi_item_part_lens, token_pos_in_items_raw_lens)
+                    multi_item_part_lens, token_pos_in_items_raw_lens, strict=True)
                 for item_len in (
-                    req_part_lens[1:] +
-                    [token_pos_in_items_len - token_pos_in_items_raw_len])
+                    [*req_part_lens[1:],
+                     token_pos_in_items_len - token_pos_in_items_raw_len])
             ],

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@tensorrt_llm/_torch/attention_backend/flashinfer.py` around lines 766 - 770,
In the nested list comprehension around the zip() call, add the strict=True
parameter to the zip() function that iterates over multi_item_part_lens and
token_pos_in_items_raw_lens to ensure both iterables have the same length.
Additionally, replace the list concatenation operation that combines
req_part_lens[1:] with a single-element list containing the computation
(token_pos_in_items_len - token_pos_in_items_raw_len) with iterable unpacking
syntax for improved clarity.

Source: Linters/SAST tools

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Inline comments:
In `@tensorrt_llm/_torch/pyexecutor/model_engine.py`:
- Around line 4319-4321: The assert statement checking multi_item_part_lens is
None causes a hard failure (AssertionError) when multi-item scoring is requested
on the CUDA graph path, but eager execution already supports this input. Instead
of asserting and aborting the request, implement a fallback mechanism: when
multi_item_part_lens is present in the graph execution path, either route the
computation to non-graph (eager) execution, or disable graph selection upstream
before this code point is reached. Replace the assert statement with conditional
logic that gracefully handles the multi-item case without raising an exception.

---

Nitpick comments:
In `@tensorrt_llm/_torch/attention_backend/flashinfer.py`:
- Around line 766-770: In the nested list comprehension around the zip() call,
add the strict=True parameter to the zip() function that iterates over
multi_item_part_lens and token_pos_in_items_raw_lens to ensure both iterables
have the same length. Additionally, replace the list concatenation operation
that combines req_part_lens[1:] with a single-element list containing the
computation (token_pos_in_items_len - token_pos_in_items_raw_len) with iterable
unpacking syntax for improved clarity.

🪄 Autofix (Beta)

Fix all unresolved CodeRabbit comments on this PR:

Push a commit to this branch (recommended)
Create a new PR with the fixes

ℹ️ Review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Enterprise

Run ID: 79445b4c-c3eb-454c-9dc2-d90ff60b1a42

📥 Commits

Reviewing files that changed from the base of the PR and between 0b0a03e and bc941ee.

📒 Files selected for processing (8)

tensorrt_llm/_torch/attention_backend/flashinfer.py
tensorrt_llm/_torch/attention_backend/interface.py
tensorrt_llm/_torch/attention_backend/star_flashinfer.py
tensorrt_llm/_torch/attention_backend/trtllm.py
tensorrt_llm/_torch/attention_backend/vanilla.py
tensorrt_llm/_torch/modules/attention.py
tensorrt_llm/_torch/pyexecutor/model_engine.py
tensorrt_llm/llmapi/llm.py

💤 Files with no reviewable changes (1)

tensorrt_llm/_torch/modules/attention.py

tensorrt-cicd · 2026-06-16T10:31:20Z

PR_Github #54575 [ run ] triggered by Bot. Commit: bc941ee Link to invocation

Funatiq

Runtime changes lgtm.

ixlmar · 2026-06-16T11:48:25Z

/bot run --disable-fail-fast

tensorrt-cicd · 2026-06-16T11:55:09Z

PR_Github #54578 [ run ] triggered by Bot. Commit: 800c7ee Link to invocation

tensorrt-cicd · 2026-06-16T11:55:13Z

PR_Github #54575 [ run ] completed with state ABORTED. Commit: bc941ee

Link to invocation

ixlmar · 2026-06-16T11:59:37Z

Note: SGLang is employing a similar optimization (sgl-project/sglang#10979)

ixlmar · 2026-06-16T20:21:02Z

/bot run --disable-fail-fast

tensorrt-cicd · 2026-06-16T20:27:21Z

PR_Github #54664 [ run ] triggered by Bot. Commit: 800c7ee Link to invocation

tensorrt-cicd · 2026-06-16T20:27:27Z

PR_Github #54578 [ run ] completed with state ABORTED. Commit: 800c7ee
/LLM/main/L0_MergeRequest_PR pipeline #43621 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 · 2026-06-16T22:41:50Z

PR_Github #54664 [ run ] completed with state SUCCESS. Commit: 800c7ee
/LLM/main/L0_MergeRequest_PR pipeline #43696 completed with status: 'SUCCESS'
Pipeline passed with automatic retried tests. Check the rerun report for details.

CI Report

Link to invocation

ixlmar · 2026-06-17T06:18:32Z

/bot run --disable-fail-fast

tensorrt-cicd · 2026-06-17T06:24:09Z

PR_Github #54776 [ run ] triggered by Bot. Commit: 1110424 Link to invocation

ixlmar · 2026-06-23T10:11:45Z

/bot run --disable-fail-fast

ixlmar · 2026-06-23T10:20:36Z

/bot run --disable-fail-fast

tensorrt-cicd · 2026-06-23T10:26:33Z

PR_Github #55225 [ run ] triggered by Bot. Commit: 26501b2 Link to invocation

tensorrt-cicd · 2026-06-23T10:57:59Z

PR_Github #55225 [ run ] completed with state FAILURE. Commit: 26501b2
/LLM/main/L0_MergeRequest_PR pipeline #44184 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

ixlmar · 2026-06-23T15:07:46Z

/bot run --disable-fail-fast

tensorrt-cicd · 2026-06-23T15:13:43Z

PR_Github #55252 [ run ] triggered by Bot. Commit: 26501b2 Link to invocation

tensorrt-cicd · 2026-06-23T15:41:31Z

PR_Github #55252 [ run ] completed with state FAILURE. Commit: 26501b2
/LLM/main/L0_MergeRequest_PR pipeline #44211 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

ixlmar · 2026-06-23T16:11:03Z

/bot run --disable-fail-fast

ixlmar · 2026-06-23T16:13:04Z

/bot run --stage-list "GB200-4GPUs-PyTorch-PerfSanity-1,GB200-4GPUs-PyTorch-PerfSanity-2"

ixlmar · 2026-06-23T16:14:04Z

/bot help

github-actions · 2026-06-23T16:14:20Z

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) --high-priority]

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. Supports wildcard * for pattern matching (e.g., "*PerfSanity*" matches all stages containing PerfSanity). Examples: "A10-PyTorch-1, xxx", "PerfSanity". 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. Supports wildcard * for pattern matching. Examples: --extra-stage "H100_PCIe-TensorRT-Post-Merge-1, xxx", --extra-stage "Post-Merge".

--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.

--high-priority (OPTIONAL) : Run the pipeline with high priority. This option is restricted to authorized users only and will route the job to a high-priority queue.

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.

tensorrt-cicd · 2026-06-23T16:17:48Z

PR_Github #55266 [ run ] triggered by Bot. Commit: f443965 Link to invocation

tensorrt-cicd · 2026-06-23T16:18:46Z

PR_Github #55267 [ run ] triggered by Bot. Commit: f443965 Link to invocation

tensorrt-cicd · 2026-06-23T16:18:48Z

PR_Github #55266 [ run ] completed with state ABORTED. Commit: f443965

Link to invocation

ixlmar · 2026-06-23T16:36:22Z

/bot run --stage-list "GB200-4GPUs-PyTorch-PerfSanity-1,GB200-4GPUs-PyTorch-PerfSanity-2"

tensorrt-cicd · 2026-06-23T16:41:40Z

PR_Github #55267 [ run ] completed with state FAILURE. Commit: f443965
/LLM/main/L0_MergeRequest_PR pipeline #44221 (Partly Tested) 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 · 2026-06-23T16:42:23Z

PR_Github #55273 [ run ] triggered by Bot. Commit: f443965 Link to invocation

tensorrt-cicd · 2026-06-23T17:17:29Z

PR_Github #55273 [ run ] completed with state FAILURE. Commit: f443965
/LLM/main/L0_MergeRequest_PR pipeline #44227 (Partly Tested) 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

ixlmar · 2026-06-24T06:23:11Z

/bot run --stage-list "GB200-4_GPUs-PyTorch-PerfSanity-1,GB200-4_GPUs-PyTorch-PerfSanity-2"

tensorrt-cicd · 2026-06-24T06:28:52Z

PR_Github #55426 [ run ] triggered by Bot. Commit: f443965 Link to invocation

ixlmar · 2026-06-24T06:38:47Z

/bot run --stage-list "GB200-4_GPUs-PyTorch-PerfSanity-1,GB200-4_GPUs-PyTorch-PerfSanity-2" --disable-fail-fast

tensorrt-cicd · 2026-06-24T06:44:15Z

PR_Github #55426 [ run ] completed with state FAILURE. Commit: f443965
/LLM/main/L0_MergeRequest_PR pipeline #44364 (Partly Tested) 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

tensorrt-cicd · 2026-06-24T06:44:48Z

PR_Github #55429 [ run ] triggered by Bot. Commit: f443965 Link to invocation

tensorrt-cicd · 2026-06-24T09:02:14Z

PR_Github #55429 [ run ] completed with state SUCCESS. Commit: f443965
/LLM/main/L0_MergeRequest_PR pipeline #44367 (Partly Tested) completed with status: 'SUCCESS'

CI Report

Link to invocation

ixlmar · 2026-06-24T09:09:21Z

Around 2026-06-23 10:00 CEST, all but two stages had passed. At the time, the branch had last been updated around 2026-06-22 20:00 CEST, the corresponding commit is now less than 50 commits behind main. The two pending stages were blocked by infra issues, the fixing of which required another branch update. The pending stages completed successfully in the latest CI run.

ixlmar · 2026-06-24T09:09:37Z

/bot skip --comment "#15413 (comment)"

tensorrt-cicd · 2026-06-24T09:15:12Z

PR_Github #55472 [ skip ] triggered by Bot. Commit: f443965 Link to invocation

tensorrt-cicd · 2026-06-24T09:24:47Z

PR_Github #55472 [ skip ] completed with state SUCCESS. Commit: f443965
Skipping testing for commit f443965

Link to invocation

ixlmar requested a review from Funatiq June 16, 2026 10:24

ixlmar requested review from a team as code owners June 16, 2026 10:24

ixlmar requested review from PerkzZheng and hchings June 16, 2026 10:24

github-actions Bot assigned ixlmar Jun 16, 2026

coderabbitai Bot reviewed Jun 16, 2026

View reviewed changes

Comment thread tensorrt_llm/_torch/pyexecutor/model_engine.py Outdated

Funatiq approved these changes Jun 16, 2026

View reviewed changes

Comment thread tensorrt_llm/_torch/pyexecutor/model_engine.py Outdated

ixlmar mentioned this pull request Jun 16, 2026

[TRTLLM-12982][chore] relocate torch_multi_arange #15416

Open

1 task

ixlmar force-pushed the perf/multi-item-scoring branch from 800c7ee to 1110424 Compare June 17, 2026 06:18

yuxianq reviewed Jun 17, 2026

View reviewed changes

Comment thread tensorrt_llm/_torch/attention_backend/interface.py Outdated

litaotju reviewed Jun 17, 2026

View reviewed changes

Comment thread tensorrt_llm/llmapi/llm.py Outdated

yuxianq reviewed Jun 17, 2026

View reviewed changes

Comment thread tensorrt_llm/llmapi/llm.py Outdated

yuxianq reviewed Jun 17, 2026

View reviewed changes

Comment thread tensorrt_llm/_torch/pyexecutor/model_engine.py

yuxianq reviewed Jun 17, 2026

View reviewed changes

Comment thread tensorrt_llm/_torch/attention_backend/flashinfer.py Outdated

ixlmar removed the request for review from a team June 17, 2026 09:30

Merge branch 'main' into perf/multi-item-scoring

f443965

ixlmar merged commit 2708009 into NVIDIA:main Jun 24, 2026
7 checks passed

Conversation

ixlmar commented Jun 16, 2026 • edited by coderabbitai Bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Description

Test Coverage

PR Checklist

GitHub Bot Help

Summary by CodeRabbit

Uh oh!

ixlmar commented Jun 16, 2026

Uh oh!

coderabbitai Bot commented Jun 16, 2026

Walkthrough

Changes

Estimated code review effort

Possibly related PRs

Suggested reviewers

Uh oh!

coderabbitai Bot left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

tensorrt-cicd commented Jun 16, 2026

Uh oh!

Funatiq left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

ixlmar commented Jun 16, 2026

Uh oh!

tensorrt-cicd commented Jun 16, 2026

Uh oh!

tensorrt-cicd commented Jun 16, 2026

Uh oh!

ixlmar commented Jun 16, 2026

Uh oh!

ixlmar commented Jun 16, 2026

Uh oh!

tensorrt-cicd commented Jun 16, 2026

Uh oh!

tensorrt-cicd commented Jun 16, 2026

Uh oh!

tensorrt-cicd commented Jun 16, 2026

Uh oh!

ixlmar commented Jun 17, 2026

Uh oh!

tensorrt-cicd commented Jun 17, 2026

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

ixlmar commented Jun 23, 2026

Uh oh!

ixlmar commented Jun 23, 2026

Uh oh!

tensorrt-cicd commented Jun 23, 2026

Uh oh!

tensorrt-cicd commented Jun 23, 2026

Uh oh!

ixlmar commented Jun 23, 2026

Uh oh!

tensorrt-cicd commented Jun 23, 2026

Uh oh!

tensorrt-cicd commented Jun 23, 2026

Uh oh!

ixlmar commented Jun 23, 2026

Uh oh!

ixlmar commented Jun 23, 2026

Uh oh!

ixlmar commented Jun 23, 2026

Uh oh!

github-actions Bot commented Jun 23, 2026

GitHub Bot Help

kill

skip

reuse-pipeline

Uh oh!

tensorrt-cicd commented Jun 23, 2026

ixlmar commented Jun 16, 2026 •

edited by coderabbitai Bot

Loading