Implementation:Openai Openai python Images Response Model

Knowledge Sources	openai-python OpenAI Image API
Domains	Computer_Vision, Data_Processing
Last Updated	2026-02-15 00:00 GMT

Overview

Concrete typed response models for extracting image generation results and creating image variations provided by the OpenAI Python SDK.

Description

ImagesResponse contains a list of Image objects with URL, base64 data, and revised prompt. The Images.create_variation() method generates variations of existing images. Both models provide typed access to all result fields.

Usage

Access response.data[0].url or response.data[0].b64_json for image data. Check revised_prompt for DALL-E 3's interpretation. Use create_variation() to generate image variations.

Code Reference

Source Location

• Repository: openai-python
• File: src/openai/types/images_response.py
• Lines: L1-79
• File: src/openai/types/image.py
• Lines: L1-28
• File: src/openai/resources/images.py
• Lines: L47-118 (create_variation)

Signature

class ImagesResponse(BaseModel):
    created: int
    data: List[Image]
    usage: Optional[ImagesResponseUsage] = None

class Image(BaseModel):
    b64_json: Optional[str] = None
    revised_prompt: Optional[str] = None
    url: Optional[str] = None

class Images(SyncAPIResource):
    def create_variation(
        self,
        *,
        image: FileTypes,
        model: Union[str, ImageModel, None] | NotGiven = NOT_GIVEN,
        n: Optional[int] | NotGiven = NOT_GIVEN,
        response_format: Optional[Literal["url", "b64_json"]] | NotGiven = NOT_GIVEN,
        size: Optional[Literal["256x256", "512x512", "1024x1024"]] | NotGiven = NOT_GIVEN,
        user: str | NotGiven = NOT_GIVEN,
    ) -> ImagesResponse:
        """Creates a variation of a given image."""

Import

from openai.types import ImagesResponse, Image
from openai import OpenAI
# create_variation via client.images.create_variation()

I/O Contract

Inputs

Name	Type	Required	Description
response	ImagesResponse	Yes	API response from generate/edit
image (variation)	FileTypes	Yes	Source image for variation

Outputs

Name	Type	Description
data[i].url	str or None	Temporary image URL (valid ~60 min)
data[i].b64_json	str or None	Base64-encoded image data
data[i].revised_prompt	str or None	DALL-E 3 revised prompt
usage	ImagesResponseUsage or None	Token/image usage stats

Usage Examples

Extract and Save Images

import base64
from openai import OpenAI

client = OpenAI()
response = client.images.generate(
    model="gpt-image-1",
    prompt="A serene Japanese garden",
    response_format="b64_json",
    n=2,
)

for i, image in enumerate(response.data):
    image_bytes = base64.b64decode(image.b64_json)
    with open(f"garden_{i}.png", "wb") as f:
        f.write(image_bytes)

Create Variations

response = client.images.create_variation(
    image=open("original.png", "rb"),
    n=4,
    size="1024x1024",
)
for image in response.data:
    print(image.url)

Related Pages

Implements Principle

• Principle:Openai_Openai_python_Image_Result_Processing

Requires Environment

• Environment:Openai_Openai_python_Python_3_9_Plus