[https://nvbugs/5449155][fix] Fix DeepSeek R1 weight loading for TP16

[achartier]

Summary by CodeRabbit

• Bug Fixes
  • Improved compatibility with FP8 weight-scale tensors by properly handling 4D singleton-friendly shapes in linear layers.
  • Fixed shape errors when loading weights for fused Q/K/V projections by adjusting per-scale handling before concatenation.
  • Resolved scale handling for fused gate/up projections to avoid extra squeezing after concatenation.
  • Preserves public APIs while ensuring smoother model loading with FP8 block scales.

[https://nvbugs/5449155][fix] Fix DeepSeek R1 weight loading for TP16

Description

Fix regression introduced by #6106

The weight_scale tensor second dimension can be equal to 1 for high TP values and get squeezed away. Replace the unconditional squeeze by one checking for the modelopt format of 4 dimensions and squeezing only the extra dimensions.

Also do the reshaping prior to calling load_weight_shard as the slice depending on TP rank uses dimension 1 for the split dimension when tp_mode is COLUMN. load_weights_fused_qkv_linear and load_weights_fused_gate_up_linear have tp_mode equal to ROW, so the change is not strictly required, but done for the sake of consistency.

Test Coverage

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.

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

Adjusts FP8BlockScalesLinearMethod internal handling of weight-scale tensors to support 4D shapes by conditionally squeezing dims 1 and -1 before shard loading and removing post-concatenation squeezes in vanilla, fused QKV, and fused gate-up loaders. No public API changes.

Changes

Cohort / File(s)

Summary

FP8 FP-scale handling updates tensorrt_llm/_torch/modules/linear.py

- load_weights_vanilla: detect 4D weight_scale and squeeze dims 1 and -1 prior to load_weight_shard; remove final squeeze. - load_weights_fused_qkv_linear: create full_q_scale, full_k_scale, full_v_scale; squeeze dims 1 and -1 if 4D; load shards and concatenate without final squeeze. - load_weights_fused_gate_up_linear: create full_left_scale, full_right_scale; squeeze dims 1 and -1 if 4D; load shards and concatenate without final squeeze.

Sequence Diagram(s)

sequenceDiagram
  participant Caller
  participant FP8Method as FP8BlockScalesLinearMethod
  participant Sharder as load_weight_shard

  Caller->>FP8Method: load_weights_*(weights, weight_scales)
  alt scale is 4D
    FP8Method->>FP8Method: squeeze dims 1 and -1 of scale
  end
  FP8Method->>Sharder: load_weight_shard(squeezed_or_original_scale)
  Sharder-->>FP8Method: shard(s)
  FP8Method->>FP8Method: concatenate shards (no post-concat squeeze)
  FP8Method-->>Caller: loaded weights / scales

Loading

Estimated code review effort

🎯 3 (Moderate) | ⏱️ ~15 minutes

Suggested reviewers

• yuxianq
• kaiyux
• hlu1
• yilin-void

Tip

🔌 Remote MCP (Model Context Protocol) integration is now available!

Pro plan users can now connect to remote MCP servers from the Integrations page. Connect with popular remote MCPs such as Notion and Linear to add more context to your reviews and chats.

---

📜 Recent review details

Configuration used: .coderabbit.yaml
Review profile: CHILL
Plan: Pro

💡 Knowledge Base configuration:

• MCP integration is disabled by default for public repositories
• Jira integration is disabled by default for public repositories
• Linear integration is disabled by default for public repositories

You can enable these sources in your CodeRabbit configuration.

📥 Commits

Reviewing files that changed from the base of the PR and between 2eb5595 and 1676f3b.

📒 Files selected for processing (1)

• tensorrt_llm/_torch/modules/linear.py (3 hunks)

🚧 Files skipped from review as they are similar to previous changes (1)

• tensorrt_llm/_torch/modules/linear.py

⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (1)

• GitHub Check: Pre-commit Check

✨ Finishing Touches

• 📝 Generate Docstrings

🧪 Generate unit tests

• Create PR with unit tests
• Post copyable unit tests in a comment

---

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

🪧 Tips

Chat

There are 3 ways to chat with CodeRabbit:

• Review comments: Directly reply to a review comment made by CodeRabbit. Example:
  • I pushed a fix in commit <commit_id>, please review it.
  • Open a follow-up GitHub issue for this discussion.
• Files and specific lines of code (under the "Files changed" tab): Tag @coderabbitai in a new review comment at the desired location with your query.
• PR comments: Tag @coderabbitai in a new PR comment to ask questions about the PR branch. For the best results, please provide a very specific query, as very limited context is provided in this mode. Examples:
  • @coderabbitai gather interesting stats about this repository and render them as a table. Additionally, render a pie chart showing the language distribution in the codebase.
  • @coderabbitai read the files in the src/scheduler package and generate a class diagram using mermaid and a README in the markdown format.

Support

Need help? Create a ticket on our support page for assistance with any issues or questions.

CodeRabbit Commands (Invoked using PR/Issue comments)

Type @coderabbitai help to get the list of available commands.

Other keywords and placeholders

• Add @coderabbitai ignore anywhere in the PR description to prevent this PR from being reviewed.
• Add @coderabbitai summary to generate the high-level summary at a specific location in the PR description.
• Add @coderabbitai or @coderabbitai title anywhere in the PR title to generate the title automatically.

Status, Documentation and Community

• Visit our Status Page to check the current availability of CodeRabbit.
• Visit our Documentation for detailed information on how to use CodeRabbit.
• Join our Discord Community to get help, request features, and share feedback.
• Follow us on X/Twitter for updates and announcements.

[coderabbitai]

Actionable comments posted: 3

📜 Review details

Configuration used: .coderabbit.yaml
Review profile: CHILL
Plan: Pro

💡 Knowledge Base configuration:

• MCP integration is disabled by default for public repositories
• Jira integration is disabled by default for public repositories
• Linear integration is disabled by default for public repositories

You can enable these settings in your CodeRabbit configuration.

📥 Commits

Reviewing files that changed from the base of the PR and between a8618b2 and 47d7468.

📒 Files selected for processing (1)

• tensorrt_llm/_torch/modules/linear.py (3 hunks)

🧰 Additional context used

📓 Path-based instructions (2)

**/*.py

📄 CodeRabbit Inference Engine (CODING_GUIDELINES.md)

**/*.py: Python code must target Python 3.8+
Python indentation: 4 spaces, no tabs
Maintain module namespace in imports (from package.subpackage import foo; then use foo.SomeClass())
Python file names use snake_case
Python class names use PascalCase
Python functions/methods and local variables use snake_case; variables starting with a number get k_ prefix (e.g., k_99th_percentile)
Global variables use G_ prefixed UPPER_SNAKE_CASE (e.g., G_MY_GLOBAL)
Constants use UPPER_SNAKE_CASE in Python
Avoid shadowing variables from outer scopes in Python
Initialize all externally visible members of a Python class in init
Prefer docstrings for interfaces used outside a file; comments for local code
Use Google-style docstrings for classes and functions (Sphinx-parsable)
Document attributes/variables inline with short docstrings
Avoid reflection when simple alternatives exist (e.g., prefer explicit parameters over dict(**locals()))
In try/except, catch the narrowest exceptions possible
For duck-typing with try/except, keep try body minimal and put logic in else

Files:

• tensorrt_llm/_torch/modules/linear.py

**/*.{cpp,cxx,cc,cu,h,hpp,hxx,hh,cuh,py}

📄 CodeRabbit Inference Engine (CODING_GUIDELINES.md)

Prepend NVIDIA copyright header (current year) to all source files

Files:

• tensorrt_llm/_torch/modules/linear.py

[achartier]

/bot run

[tensorrt-cicd]

PR_Github #15381 [ run ] triggered by Bot

[tensorrt-cicd]

PR_Github #15381 [ run ] completed with state FAILURE
/LLM/release-1.0/L0_MergeRequest_PR pipeline #132 completed with status: 'FAILURE'

[achartier]

/bot run

[tensorrt-cicd]

PR_Github #15451 [ run ] triggered by Bot

[tensorrt-cicd]

PR_Github #15451 [ run ] completed with state SUCCESS
/LLM/release-1.0/L0_MergeRequest_PR pipeline #144 completed with status: 'FAILURE'

[achartier]

/bot run

[tensorrt-cicd]

PR_Github #15471 [ run ] triggered by Bot

[achartier]

/bot run

[tensorrt-cicd]

PR_Github #15471 [ run ] completed with state SUCCESS
/LLM/release-1.0/L0_MergeRequest_PR pipeline #150 completed with status: 'FAILURE'

[tensorrt-cicd]

PR_Github #15486 [ run ] triggered by Bot

[tensorrt-cicd]

PR_Github #15486 [ run ] completed with state SUCCESS
/LLM/release-1.0/L0_MergeRequest_PR pipeline #153 completed with status: 'FAILURE'

[thorjohnsen]

/bot run

[tensorrt-cicd]

PR_Github #15662 [ run ] triggered by Bot

[tensorrt-cicd]

PR_Github #15662 [ run ] completed with state SUCCESS
/LLM/release-1.0/L0_MergeRequest_PR pipeline #187 completed with status: 'SUCCESS'