Implementation:Openai Openai python Compacted Response

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

Overview

Concrete type for representing a compacted response provided by the openai-python SDK.

Description

CompactedResponse is a Pydantic model class that represents a compacted conversation response from the OpenAI API. It extends BaseModel and includes a unique identifier, a Unix timestamp for creation, an object type field fixed to "response.compaction", a list of ResponseOutputItem objects representing the compacted output (all user messages followed by a single compaction item), and a ResponseUsage object providing token accounting for the compaction pass including cached, reasoning, and total tokens.

Usage

Import this type when working with conversation compaction in the Responses API. The compacted response is returned when a long conversation is condensed to reduce token usage while preserving context.

Code Reference

Source Location

• Repository: openai-python
• File: src/openai/types/responses/compacted_response.py

Signature

class CompactedResponse(BaseModel):
    id: str
    created_at: int
    object: Literal["response.compaction"]
    output: List[ResponseOutputItem]
    usage: ResponseUsage

Import

from openai.types.responses import CompactedResponse

I/O Contract

Fields

Name	Type	Required	Description
id	str	Yes	The unique identifier for the compacted response.
created_at	int	Yes	Unix timestamp (in seconds) when the compacted conversation was created.
object	Literal["response.compaction"]	Yes	The object type. Always "response.compaction".
output	List[ResponseOutputItem]	Yes	The compacted list of output items (user messages followed by a compaction item).
usage	ResponseUsage	Yes	Token accounting for the compaction pass, including cached, reasoning, and total tokens.

Usage Examples

from openai.types.responses import CompactedResponse

# Access fields from a compacted response
compacted: CompactedResponse = ...
print(f"Compaction ID: {compacted.id}")
print(f"Created at: {compacted.created_at}")
print(f"Output items: {len(compacted.output)}")
print(f"Total tokens: {compacted.usage.total_tokens}")

Related Pages

• Environment:Openai_Openai_python_Python_3_9_Plus