Providers and ModelsOpenAI

GPT Image 2

GPT Image 2
OpenAI · Image Generation
POST /v1/images/generations

OpenAI’s flagship image model with strong prompt fidelity, crisp text rendering, and instruction-based editing across up to 16 reference images.

At a glance

FieldValue
Model idgpt-image-2
Model release date2026-04-21
Input modalitiesText, Image
Output modalitiesImage
Context window-
Weight precision-
Featuresimage_generation, image_editing, text_rendering, multi_image
Native inferenceNo
NewNo
Supported endpointsPOST /v1/images/generations
Alternate model idsopenai/gpt-image-2

Pricing

ChargeSpecRate
Lowper image$0.012
Mediumper image$0.106
Highper image$0.422

Example request

$curl https://api.empiriolabs.ai/v1/images/generations \
>  -H 'Authorization: Bearer $EMPIRIOLABS_API_KEY' \
>  -H 'Content-Type: application/json' \
>  -d '{"model": "gpt-image-2", "prompt": "a misty forest at dawn", "size": "1024x1024"}'

Parameters

ParameterTypeRequiredDefaultDescription
promptstringyes-Text prompt. With input images attached, becomes an editing or compositing instruction.
qualityenumno"medium"Rendering quality tier. Higher tiers produce more detail and cost more per image. · Allowed: low, medium, high
sizeenumno"1024x1024"Output resolution: square, landscape, or portrait. · Allowed: 1024x1024, 1536x1024, 1024x1536
imagearrayno-Up to 16 input image URLs for editing or multi-reference composition.
num_imagesnumberno1Number of images to generate. Each is billed separately. · Range: 1 – 4
moderationenumno"auto"Content filtering strictness applied by the model service. · Allowed: auto, low
response_formatenumno"url"Return signed URLs or base64 image data. · Allowed: url, b64_json

Notes

Quality tiers Low, Medium, and High set the rendering fidelity and the per-image price. Output sizes are 1024x1024, 1536x1024, and 1024x1536. Attach up to 16 images to edit or composite them, with the prompt as the instruction.

Content safety

  • Generated images pass an automated content-safety check. An image flagged as explicit is not returned. Because the image is generated before the check runs, a flagged generation is still billed.

Machine-readable schema: GET https://api.empiriolabs.ai/v1/models/gpt-image-2.