Implementation:Openai Openai python Shared Reasoning

Knowledge Sources	Openai_Openai_python OpenAI API Reference
Domains	API_Types, Python
Last Updated	2026-02-15 00:00 GMT

Overview

Concrete tool for configuring reasoning behavior for gpt-5 and o-series models provided by the openai-python SDK.

Description

Reasoning is a Pydantic BaseModel that provides configuration options for reasoning models. It contains three optional fields:

• effort - A ReasoningEffort value constraining reasoning effort. Supported values are "none", "minimal", "low", "medium", "high", and "xhigh". Different models have different defaults: gpt-5.1 defaults to "none", models before gpt-5.1 default to "medium", and gpt-5-pro defaults to (and only supports) "high". The "xhigh" level is available for models after gpt-5.1-codex-max.
• generate_summary - Deprecated in favor of summary. Controls whether the model provides a summary of its reasoning process.
• summary - Controls reasoning summary output with values "auto", "concise", or "detailed". Useful for debugging and understanding the model's reasoning process.

Usage

Import Reasoning when configuring reasoning parameters for chat completions or responses API calls with reasoning-capable models.

Code Reference

Source Location

• Repository: openai-python
• File: src/openai/types/shared/reasoning.py

Signature

class Reasoning(BaseModel):
    effort: Optional[ReasoningEffort] = None
    generate_summary: Optional[Literal["auto", "concise", "detailed"]] = None
    summary: Optional[Literal["auto", "concise", "detailed"]] = None

Import

from openai.types.shared import Reasoning

I/O Contract

Fields

Name	Type	Required	Description
effort	Optional[ReasoningEffort]	No	Constrains reasoning effort. Values: "none", "minimal", "low", "medium", "high", "xhigh". Reducing effort results in faster responses and fewer reasoning tokens.
generate_summary	Optional[Literal["auto", "concise", "detailed"]]	No	(Deprecated) Use summary instead. Summary of reasoning performed by the model.
summary	Optional[Literal["auto", "concise", "detailed"]]	No	Summary of reasoning performed by the model. Useful for debugging. "concise" supported for computer-use-preview and reasoning models after gpt-5.

Usage Examples

from openai import OpenAI
from openai.types.shared import Reasoning

client = OpenAI()

# Configure reasoning with high effort and concise summary
response = client.chat.completions.create(
    model="o3",
    messages=[{"role": "user", "content": "Solve this math problem: ..."}],
    reasoning=Reasoning(
        effort="high",
        summary="concise",
    ).model_dump(),
)

# Use low effort for faster responses
response = client.chat.completions.create(
    model="gpt-5.1",
    messages=[{"role": "user", "content": "Simple question"}],
    reasoning=Reasoning(effort="low").model_dump(),
)

Related Pages

• Environment:Openai_Openai_python_Python_3_9_Plus