docs: update TensorRT-LLM commit for NIXL to v1.2.0rc2

[dagil-nvidia]

docs: update TensorRT-LLM commit to v1.2.0rc2

Summary

Updates the TensorRT-LLM commit reference from main to v1.2.0rc2 in the documentation for NIXL KV cache transfer and EPD (Encode-Prefill-Decode) flow features.

Changes

• Updated docs/backends/trtllm/kv-cache-transfer.md to use --tensorrtllm-commit v1.2.0rc2
• Updated docs/backends/trtllm/multimodal_epd.md to use --tensorrtllm-commit v1.2.0rc2

Motivation

These features require a specific TensorRT-LLM version. Using main can lead to compatibility issues as the TensorRT-LLM main branch evolves. Pinning to v1.2.0rc2 ensures users build with the correct tested version.

Testing

• Documentation builds without warnings
• No linter errors

Checklist

• Commit message follows conventional commit format (docs: prefix)
• DCO sign-off included (-s)
• Changes are documentation only
• GPG signing (note: GPG was not available during commit, can be amended if needed for automated tests)

PR Link

Create PR from dagil/fix-trtllm-commit-docs → main
Branch pushed to: https://github.com/ai-dynamo/dynamo/pull/new/dagil/fix-trtllm-commit-docs

Summary by CodeRabbit

• Documentation
  • Updated TensorRT-LLM backend build instructions to reference a specific version (v1.2.0rc2) instead of the main branch for improved consistency and reproducibility.

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

[copy-pr-bot]

This pull request requires additional validation before any workflows can run on NVIDIA's runners.

Pull request vetters can view their responsibilities here.

Contributors can view more details about this message here.

[dagil-nvidia]

/ok to test 620889e

[coderabbitai]

Walkthrough

Two TensorRT-LLM backend documentation files have been updated to pin the commit version from main branch to v1.2.0rc2 in build instructions and example commands.

Changes

Cohort / File(s)	Summary
TensorRT-LLM Build Instructions docs/backends/trtllm/kv-cache-transfer.md, docs/backends/trtllm/multimodal_epd.md	Updated build commands and enablement instructions to specify TensorRT-LLM commit v1.2.0rc2 instead of main branch reference

Estimated code review effort

🎯 1 (Trivial) | ⏱️ ~3 minutes

• Straightforward documentation version pinning updates
• No logic changes or code alterations
• Consistent pattern across both files

Poem

🐰 A rabbit hops with glee,
Pinning versions to v1.2!
No more "main" branch dreams so free,
Release candidate's the way to be,
Documentation's locked and true! 🔐

Pre-merge checks

✅ Passed checks (3 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title clearly and concisely summarizes the main change: updating TensorRT-LLM commit references in documentation from main to v1.2.0rc2.
Description check	✅ Passed	The description includes all required template sections (Overview, Details, Where should reviewer start, Related Issues) with comprehensive information about changes, motivation, and testing.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.

---

📜 Recent review details

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

📥 Commits

Reviewing files that changed from the base of the PR and between 4babb33 and 620889e.

📒 Files selected for processing (2)

• docs/backends/trtllm/kv-cache-transfer.md (1 hunks)
• docs/backends/trtllm/multimodal_epd.md (1 hunks)

🧰 Additional context used

🧠 Learnings (3)

📚 Learning: 2025-07-30T00:34:35.810Z

Learnt from: dmitry-tokarev-nv
Repo: ai-dynamo/dynamo PR: 2179
File: docs/support_matrix.md:61-63
Timestamp: 2025-07-30T00:34:35.810Z
Learning: In docs/support_matrix.md, the NIXL version difference between runtime dependencies (0.5.0) and build dependencies (0.4.0) is intentional and expected, not an error that needs to be corrected.

Applied to files:

• docs/backends/trtllm/kv-cache-transfer.md

📚 Learning: 2025-07-22T10:22:28.972Z

Learnt from: ptarasiewiczNV
Repo: ai-dynamo/dynamo PR: 2027
File: container/deps/vllm/install_vllm.sh:0-0
Timestamp: 2025-07-22T10:22:28.972Z
Learning: The `--torch-backend=auto` flag works with vLLM installations via uv pip install, even though it's not a standard pip option. This flag is processed by vLLM's build system during installation to automatically match PyTorch distribution with container CUDA versions.

Applied to files:

• docs/backends/trtllm/multimodal_epd.md

📚 Learning: 2025-08-30T20:43:10.091Z

Learnt from: keivenchang
Repo: ai-dynamo/dynamo PR: 2797
File: .devcontainer/devcontainer.json:12-12
Timestamp: 2025-08-30T20:43:10.091Z
Learning: In the dynamo project, devcontainer.json files use templated container names (like "dynamo-vllm-devcontainer") that are automatically processed by the copy_devcontainer.sh script to generate framework-specific configurations with unique names, preventing container name collisions.

Applied to files:

• docs/backends/trtllm/multimodal_epd.md

⏰ 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). (2)

• GitHub Check: Build Documentation
• GitHub Check: Build and Test - dynamo

🔇 Additional comments (2)

docs/backends/trtllm/kv-cache-transfer.md (1)

48-51: Pin TensorRT-LLM commit to v1.2.0rc2 for NIXL KV cache transfer.

The build command now explicitly targets v1.2.0rc2, aligning with the related update in multimodal_epd.md. This pinning ensures compatibility with the NIXL-based KV cache transfer feature and avoids issues from main branch changes.

docs/backends/trtllm/multimodal_epd.md (1)

11-11: Verified: v1.2.0rc2 is a valid TensorRT-LLM release tag.

The build command correctly pins to the v1.2.0rc2 release tag, which is confirmed as a published tag in the NVIDIA/TensorRT-LLM repository. The change is valid and properly formatted.

Tip

📝 Customizable high-level summaries are now available in beta!

You can now customize how CodeRabbit generates the high-level summary in your pull requests — including its content, structure, tone, and formatting.

• Provide your own instructions using the high_level_summary_instructions setting.
• Format the summary however you like (bullet lists, tables, multi-section layouts, contributor stats, etc.).
• Use high_level_summary_in_walkthrough to move the summary from the description to the walkthrough section.

Example instruction:

"Divide the high-level summary into five sections:

1. 📝 Description — Summarize the main change in 50–60 words, explaining what was done.
2. 📓 References — List relevant issues, discussions, documentation, or related PRs.
3. 📦 Dependencies & Requirements — Mention any new/updated dependencies, environment variable changes, or configuration updates.
4. 📊 Contributor Summary — Include a Markdown table showing contributions:
   | Contributor | Lines Added | Lines Removed | Files Changed |
5. ✔️ Additional Notes — Add any extra reviewer context.
   Keep each section concise (under 200 words) and use bullet or numbered lists for clarity."

Note: This feature is currently in beta for Pro-tier users, and pricing will be announced later.

---

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.}