add reasoning param for openai responses LLM

[tinalenguyen]

to use:

from openai.types import Reasoning

llm=openai.responses.LLM(reasoning=Reasoning(effort="low", summary="low")

Summary by CodeRabbit

• New Features
  • Language model operations now support configurable reasoning effort levels for enhanced flexibility
  • Compatible models automatically receive optimized default reasoning settings to ensure reliable performance
  • Reasoning configuration seamlessly integrates with all chat interactions
  • Users can customize reasoning behavior to meet specific performance and resource requirements

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

[coderabbitai]

📝 Walkthrough

Walkthrough

This change adds support for OpenAI's reasoning parameter to the LLM class. It introduces a new reasoning field to the options, accepts it as a constructor parameter with model-specific defaults, and propagates it through the chat API calls.

Changes

Cohort / File(s)	Summary
Reasoning Feature Integration livekit-plugins/livekit-plugins-openai/livekit/plugins/openai/responses/llm.py	Added reasoning: NotGivenOr[Reasoning] field to _LLMOptions class. Updated LLM.__init__() to accept reasoning parameter with model-specific defaults (effort="none" for gpt-5.1/5.2, effort="minimal" otherwise when supported). Integrated reasoning propagation into chat() extra parameters. Added imports for Reasoning type and _supports_reasoning_effort helper.

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~8 minutes

Poem

🐰 With reasoning now woven through the thread,
Our LLM thinks before it's said!
From "minimal" effort to "none" so fine,
The rabbit hops through logic's design! ✨

🚥 Pre-merge checks | ✅ 2 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 50.00% which is insufficient. The required threshold is 80.00%.	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (2 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The PR title 'add reasoning param for openai responses LLM' clearly and directly summarizes the main change: adding a reasoning parameter to the OpenAI responses LLM class, which matches the changeset content.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing touches

• 📝 Generate docstrings

---

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[coderabbitai]

Actionable comments posted: 1

🤖 Fix all issues with AI agents

In
`@livekit-plugins/livekit-plugins-openai/livekit/plugins/openai/responses/llm.py`:
- Around line 77-81: The current branch in the reasoning default logic overrides
OpenAI defaults; change the logic in the block that uses
_supports_reasoning_effort and Reasoning so that if reasoning is not provided
you do NOT force "minimal" for pre-5.1 models — instead only set
Reasoning(effort="none") for models that must default to none (e.g., "gpt-5.1"),
and otherwise leave reasoning as None/omitted so the API uses its documented
default (or explicitly set "medium" only if you intentionally want to override);
update the conditional around model and _supports_reasoning_effort to reflect
this and remove the forced "minimal" assignment.

📜 Review details

Configuration used: Organization UI

Review profile: CHILL

Plan: Pro

📥 Commits

Reviewing files that changed from the base of the PR and between 853bc41 and b2c4c96.

📒 Files selected for processing (1)

• livekit-plugins/livekit-plugins-openai/livekit/plugins/openai/responses/llm.py

🧰 Additional context used

📓 Path-based instructions (1)

**/*.py

📄 CodeRabbit inference engine (AGENTS.md)

**/*.py: Format code with ruff
Run ruff linter and auto-fix issues
Run mypy type checker in strict mode
Maintain line length of 100 characters maximum
Ensure Python 3.9+ compatibility
Use Google-style docstrings

Files:

• livekit-plugins/livekit-plugins-openai/livekit/plugins/openai/responses/llm.py

🧬 Code graph analysis (1)

livekit-plugins/livekit-plugins-openai/livekit/plugins/openai/responses/llm.py (2)

livekit-plugins/livekit-plugins-openai/livekit/plugins/openai/models.py (1)

• _supports_reasoning_effort (294-301)

livekit-agents/livekit/agents/utils/misc.py (1)

• is_given (25-26)

⏰ 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). (9)

• GitHub Check: livekit-plugins-inworld
• GitHub Check: livekit-plugins-cartesia
• GitHub Check: unit-tests
• GitHub Check: livekit-plugins-groq
• GitHub Check: livekit-plugins-deepgram
• GitHub Check: livekit-plugins-elevenlabs
• GitHub Check: livekit-plugins-openai
• GitHub Check: type-check (3.13)
• GitHub Check: type-check (3.9)

🔇 Additional comments (5)

livekit-plugins/livekit-plugins-openai/livekit/plugins/openai/responses/llm.py (5)

23-23: Imports for Reasoning and model capability helper look appropriate.

Also applies to: 37-37

---

48-48: _LLMOptions now cleanly models the new reasoning option.

---

63-63: Constructor surface updated consistently.

---

91-91: Reasoning option is propagated into _LLMOptions as expected.

---

135-136: Reasoning is forwarded into request kwargs cleanly.

_{✏️ Tip: You can disable this entire section by setting review_details to false in your review settings.}

[davidzhao]

lgtm