docs: Add note about ignore_eos for MTP

[rmccorm4]

Overview:

Adds note to docs about not using ignore_eos with MTP for realistic acceptance rates.

Summary by CodeRabbit

• Documentation
  • Improved clarity and correctness in the description of Multi-Token Prediction (MTP) usage.
  • Updated latency notes to indicate noticeable initial latency for the first two inference requests.
  • Added important guidance regarding the ignore_eos parameter when using MTP to prevent unwanted outputs.

[coderabbitai]

Walkthrough

The README for the TensorRT-LLM DeepSeek R1 examples was updated to clarify Multi-Token Prediction (MTP) usage, improve formatting, correct latency notes, and add a warning about the ignore_eos parameter to prevent incorrect outputs. No code or configuration files were changed.

Changes

File(s)	Change Summary
examples/tensorrt_llm/README.md	Improved MTP description formatting, clarified latency notes, and added guidance on ignore_eos.

Sequence Diagram(s)

No sequence diagram generated as the changes are documentation-only.

Possibly related PRs

• docs: MTP + TensorRT LLM + DS R1 disaggregated example #1473: Updates README documentation to clarify MTP usage notes and latency expectations for DeepSeek R1 with TensorRT LLM, closely related to this PR's documentation clarifications.
• docs: add aggregated example turning on MTP with DeepSeek R1  #1421: Updates the README to clarify usage notes and correct formatting for MTP with DeepSeek R1, building upon the same example and notes refined in this PR.

Poem

In the README garden, a rabbit did hop,
Tidying MTP notes with a gentle plop.
"Mind your ignore_eos, or outputs go wild—
And watch those first two requests," the bunny smiled.
With spaces and warnings, the docs now gleam,
Clear as a carrot in a coder's dream! 🥕

---

🪧 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.
  • Explain this complex logic.
  • 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. Examples:
  • @coderabbitai explain this code block.
  • @coderabbitai modularize this function.
• 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 src/utils.ts and explain its main purpose.
  • @coderabbitai read the files in the src/scheduler package and generate a class diagram using mermaid and a README in the markdown format.
  • @coderabbitai help me debug CodeRabbit configuration file.

Support

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

Note: Be mindful of the bot's finite context window. It's strongly recommended to break down tasks such as reading entire modules into smaller chunks. For a focused discussion, use review comments to chat about specific files and their changes, instead of using the PR comments.

CodeRabbit Commands (Invoked using PR comments)

• @coderabbitai pause to pause the reviews on a PR.
• @coderabbitai resume to resume the paused reviews.
• @coderabbitai review to trigger an incremental review. This is useful when automatic reviews are disabled for the repository.
• @coderabbitai full review to do a full review from scratch and review all the files again.
• @coderabbitai summary to regenerate the summary of the PR.
• @coderabbitai generate sequence diagram to generate a sequence diagram of the changes in this PR.
• @coderabbitai resolve resolve all the CodeRabbit review comments.
• @coderabbitai configuration to show the current CodeRabbit configuration for the repository.
• @coderabbitai help to get help.

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 anywhere in the PR title to generate the title automatically.

Documentation and Community

• 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: 0

🧹 Nitpick comments (4)

examples/tensorrt_llm/README.md (4)

132-136: Inconsistent config path formatting
The aggregated MTP example omits the leading ./ in the path (-f configs/...), whereas all other examples use -f ./configs/.... Align this for consistency.

---

140-141: Consider reformatting the ignore_eos guidance
This bullet is quite dense. You might split it into two bullets or convert it into an admonition/note block to improve scannability.

---

237-239: Inconsistent config path formatting in multi-node section
Similar to the aggregated example, the multi-node MTP path is written as -f configs/.... Use -f ./configs/... here as well to match the rest of the docs.

---

273-274: Duplicate ignore_eos note
The same guidance about ignore_eos appears twice. Consider referencing the original note or consolidating to avoid redundancy.

📜 Review details

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

📥 Commits

Reviewing files that changed from the base of the PR and between 3363d8b and 25eb20f.

📒 Files selected for processing (1)

• examples/tensorrt_llm/README.md (3 hunks)

⏰ Context from checks skipped due to timeout of 90000ms (2)

• GitHub Check: Mirror Repository to GitLab
• GitHub Check: Build and Test - vllm