ENH add tool calling args

[gwarmstrong]

Summary by CodeRabbit

• Documentation
  • Updated tool calling examples to include additional server configuration parameters in CLI and Python API usage examples.

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

[coderabbitai]

📝 Walkthrough

Walkthrough

Documentation examples for tool calling are updated to include additional server arguments (--enable-auto-tool-choice --tool-call-parser hermes) in CLI and Python API usage examples. No functional code changes.

Changes

Cohort / File(s)	Summary
Documentation updates docs/agentic_inference/tool_calling.md	Updated multiple example code snippets to include additional server_args parameter with '--enable-auto-tool-choice --tool-call-parser hermes' configuration.

Estimated code review effort

🎯 1 (Trivial) | ⏱️ ~2 minutes

• Straightforward documentation example updates with no code logic changes
• No control flow, error handling, or functionality modifications

Pre-merge checks and finishing touches

✅ Passed checks (3 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title 'ENH add tool calling args' directly relates to the changeset, which updates documentation with additional server_args parameters for tool calling.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.

✨ Finishing touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Post copyable unit tests in a comment
• Commit unit tests in branch georgea/update-tool-calling-docs

---

📜 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 fbdece4 and 05cc622.

📒 Files selected for processing (1)

• docs/agentic_inference/tool_calling.md (5 hunks)

⏰ 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: pre-commit
• GitHub Check: unit-tests

🔇 Additional comments (1)

docs/agentic_inference/tool_calling.md (1)

54-54: Documentation enhancement for vLLM tool calling is accurate.

The added --enable-auto-tool-choice --tool-call-parser hermes arguments are current and recommended for vLLM 0.6.x. These flags are consistently applied across all runnable examples (PythonTool, custom tools, multiple tools), improving clarity and practical usability while aligning with the existing "Server Configuration" section.

Also applies to: 77-77, 274-274, 293-293, 349-349

---

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.}