GPT Image 2

OpenAI GPT Image 2 image generation model. Next-generation autoregressive + reasoning hybrid model with flexible resolution up to 4K and near-perfect text rendering.

GPT Image 2 is OpenAI’s next-generation image generation model, available through AnyFast API. Built on an autoregressive + reasoning hybrid architecture, it delivers native 2K resolution, near-perfect text rendering (~99% character-level accuracy across 12+ languages), and superior multi-object scene composition.

Key capabilities

• Text-to-Image — Generate images from natural language descriptions
• Image editing — Edit an existing image with a text prompt via /v1/images/edits
• Flexible resolution — Any custom size up to 4K (3840px), edges must be multiples of 16
• Near-perfect text rendering — ~99% character-level accuracy across 12+ languages
• Multi-object composition — Complex scenes without occlusion or misplacement issues
• Multi-style — Photorealism, illustration, anime, vector, 3D, data visualization

Both /v1/images/generations and /v1/images/edits support n=1–10 (default 1). Requires selecting the Direct group in the console.

Output specifications

Property	Value
Sizes	Flexible resolution (e.g. 1024x1024, 2048x2048, 3840x2160)
Size constraints	Edges: multiples of 16, aspect ratio ≤ 3:1, total pixels 655,360–8,294,400
Quality	low, medium, high
Output formats	png, jpeg
Input formats (edits)	png, jpeg

Quick example

cURL

curl https://www.anyfast.ai/v1/images/generations \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-image-2",
    "prompt": "A futuristic city skyline at sunset with flying cars",
    "size": "2048x2048",
    "quality": "high"
  }'

from openai import OpenAI

client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://www.anyfast.ai/v1"
)

response = client.images.generate(
    model="gpt-image-2",
    prompt="A futuristic city skyline at sunset with flying cars",
    size="2048x2048",
    quality="high"
)

print(response.data[0].b64_json[:100])

Parameters

Parameter	Type	Required	Description
model	string	Yes	Must be gpt-image-2
prompt	string	Yes	Text description of the desired image
n	integer	No	Number of images to generate. Supports 1–10. Default: 1. Requires selecting the Direct group.
size	string	No	{width}x{height} format. Edges must be multiples of 16, aspect ratio ≤ 3:1, total pixels 655,360–8,294,400, max edge 3,840px. Default: 1024x1024
quality	string	No	low, medium, high. Default: medium
output_format	string	No	png, jpeg. Default: png
moderation	string	No	auto or low. Default: auto
output_compression	integer	No	Compression level for jpeg (0–100).

Image editing

Edit an existing image with a text prompt via POST /v1/images/edits. Use multipart/form-data and pass images as file uploads.

Supports up to 16 input images per request (image[]).

Input image formats

Supported formats: PNG, JPEG. Images must be provided as multipart/form-data file uploads.

Looking for a lower-cost option with response_format support (no n parameter)? See gpt-image-2-c.

Single image

curl https://www.anyfast.ai/v1/images/edits \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -F "image=@/path/to/source.jpeg" \
  -F model="gpt-image-2" \
  -F prompt="Change the background to a sunset over the ocean" \
  -F n=1 \
  -F size="1024x1024"

curl https://www.anyfast.ai/v1/images/edits \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -F "image[]=@/path/to/image1.png" \
  -F "image[]=@/path/to/image2.png" \
  -F "image[]=@/path/to/image3.png" \
  -F model="gpt-image-2" \
  -F prompt="Generate a gift basket containing all the items in the reference images" \
  -F size="1024x1024"

Edit parameters

Parameter	Type	Required	Description
image	file	Yes	Single source image (PNG or JPEG, multipart upload)
image[]	file array	Yes*	Multiple source images — up to 16 images (PNG or JPEG). Use instead of image for multi-image input
mask	file	No	Mask image with an alpha channel. Same format and size as source (< 50MB). Applied to the first image when multiple images are provided. Requires selecting the Direct group.
model	string	Yes	Must be gpt-image-2
prompt	string	Yes	Edit instruction text
n	integer	No	Number of images to return. Supports 1–10. Default: 1. Requires selecting the Direct group.
size	string	No	{width}x{height}. Default: 1024x1024
output_format	string	No	png, jpeg. Default: png
output_compression	integer	No	Compression level for jpeg (0–100)

Use either image (single file) or image[] (array) — not both.

Mask editing

Provide a mask image with an alpha channel to control which areas of the source image to edit. Pixels with white (opaque) alpha are preserved; pixels with black (transparent) alpha are edited.

Requirements:

• The mask must contain an Alpha channel
• Must be the same image format and same dimensions as the source image
• File size limit: < 50MB

Both mask editing and the n parameter require selecting the Direct group in the console.

cURL

curl https://www.anyfast.ai/v1/images/edits \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -F image=@/path/to/source.jpeg \
  -F mask=@/path/to/mask.png \
  -F model="gpt-image-2" \
  -F prompt="Replace the background with a futuristic cityscape" \
  -F n=1 \
  -F size="2048x2048"

If your mask is a black-and-white image without an alpha channel, convert it first:

Python

from PIL import Image
import numpy as np

img = Image.open("bw_mask.png").convert("L")
alpha = np.array(img)
rgba = np.zeros((*alpha.shape, 4), dtype=np.uint8)
rgba[alpha > 0] = [255, 255, 255, 255]
Image.fromarray(rgba).save("mask.png")

Generate API Reference

Interactive playground for POST /v1/images/generations.

Edit API Reference

Interactive playground for POST /v1/images/edits.