Implementation:Openai Openai python Shared Params Compound Filter

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

Overview

Concrete tool for combining multiple filters using logical AND or OR operations in vector store search contexts, provided by the openai-python SDK.

Description

CompoundFilter is a TypedDict that allows combining multiple filters using and or or logic. It contains two required fields: filters (an iterable of ComparisonFilter or other CompoundFilter objects for recursive nesting) and type (either "and" or "or"). This enables constructing complex, nested filter expressions for vector store searches and other filtering endpoints.

Usage

Import this type when you need to combine multiple ComparisonFilter conditions with logical operators for vector store search queries or other API endpoints that accept compound filters.

Code Reference

Source Location

• Repository: openai-python
• File: src/openai/types/shared_params/compound_filter.py

Signature

Filter: TypeAlias = Union[ComparisonFilter, object]


class CompoundFilter(TypedDict, total=False):
    """Combine multiple filters using `and` or `or`."""

    filters: Required[Iterable[Filter]]
    type: Required[Literal["and", "or"]]

Import

from openai.types.shared_params import CompoundFilter

I/O Contract

Fields

Name	Type	Required	Description
filters	Iterable[Union[ComparisonFilter, object]]	Yes	Array of filters to combine. Items can be ComparisonFilter or CompoundFilter.
type	Literal["and", "or"]	Yes	Type of logical operation: "and" or "or".

Usage Examples

from openai.types.shared_params import CompoundFilter, ComparisonFilter

# Combine filters: category is "docs" AND score > 0.5
compound: CompoundFilter = {
    "type": "and",
    "filters": [
        {"key": "category", "type": "eq", "value": "docs"},
        {"key": "score", "type": "gt", "value": 0.5},
    ],
}

# Use in vector store search
from openai import OpenAI
client = OpenAI()
results = client.vector_stores.search(
    "vs_abc123",
    query="deployment guide",
    filters=compound,
)

Related Pages

• Environment:Openai_Openai_python_Python_3_9_Plus