GPT-Image-2

GPT-Image-2

curl --request POST \
  --url https://api.gravitex.ai/v1/images/generations \
  --header 'Authorization: <authorization>' \
  --header 'Content-Type: <content-type>' \
  --data '
{
  "model": "<string>",
  "prompt": "<string>",
  "image": [
    "<string>"
  ],
  "n": 123,
  "size": "<string>",
  "quality": "<string>",
  "background": "<string>",
  "output_format": "<string>",
  "output_compression": 123,
  "moderation": "<string>",
  "user": "<string>"
}
'

import requests

url = "https://api.gravitex.ai/v1/images/generations"

payload = {
    "model": "<string>",
    "prompt": "<string>",
    "image": ["<string>"],
    "n": 123,
    "size": "<string>",
    "quality": "<string>",
    "background": "<string>",
    "output_format": "<string>",
    "output_compression": 123,
    "moderation": "<string>",
    "user": "<string>"
}
headers = {
    "Authorization": "<authorization>",
    "Content-Type": "<content-type>"
}

response = requests.post(url, json=payload, headers=headers)

print(response.text)

const options = {
  method: 'POST',
  headers: {Authorization: '<authorization>', 'Content-Type': '<content-type>'},
  body: JSON.stringify({
    model: '<string>',
    prompt: '<string>',
    image: ['<string>'],
    n: 123,
    size: '<string>',
    quality: '<string>',
    background: '<string>',
    output_format: '<string>',
    output_compression: 123,
    moderation: '<string>',
    user: '<string>'
  })
};

fetch('https://api.gravitex.ai/v1/images/generations', options)
  .then(res => res.json())
  .then(res => console.log(res))
  .catch(err => console.error(err));

<?php

$curl = curl_init();

curl_setopt_array($curl, [
  CURLOPT_URL => "https://api.gravitex.ai/v1/images/generations",
  CURLOPT_RETURNTRANSFER => true,
  CURLOPT_ENCODING => "",
  CURLOPT_MAXREDIRS => 10,
  CURLOPT_TIMEOUT => 30,
  CURLOPT_HTTP_VERSION => CURL_HTTP_VERSION_1_1,
  CURLOPT_CUSTOMREQUEST => "POST",
  CURLOPT_POSTFIELDS => json_encode([
    'model' => '<string>',
    'prompt' => '<string>',
    'image' => [
        '<string>'
    ],
    'n' => 123,
    'size' => '<string>',
    'quality' => '<string>',
    'background' => '<string>',
    'output_format' => '<string>',
    'output_compression' => 123,
    'moderation' => '<string>',
    'user' => '<string>'
  ]),
  CURLOPT_HTTPHEADER => [
    "Authorization: <authorization>",
    "Content-Type: <content-type>"
  ],
]);

$response = curl_exec($curl);
$err = curl_error($curl);

curl_close($curl);

if ($err) {
  echo "cURL Error #:" . $err;
} else {
  echo $response;
}

package main

import (
	"fmt"
	"strings"
	"net/http"
	"io"
)

func main() {

	url := "https://api.gravitex.ai/v1/images/generations"

	payload := strings.NewReader("{\n  \"model\": \"<string>\",\n  \"prompt\": \"<string>\",\n  \"image\": [\n    \"<string>\"\n  ],\n  \"n\": 123,\n  \"size\": \"<string>\",\n  \"quality\": \"<string>\",\n  \"background\": \"<string>\",\n  \"output_format\": \"<string>\",\n  \"output_compression\": 123,\n  \"moderation\": \"<string>\",\n  \"user\": \"<string>\"\n}")

	req, _ := http.NewRequest("POST", url, payload)

	req.Header.Add("Authorization", "<authorization>")
	req.Header.Add("Content-Type", "<content-type>")

	res, _ := http.DefaultClient.Do(req)

	defer res.Body.Close()
	body, _ := io.ReadAll(res.Body)

	fmt.Println(string(body))

}

HttpResponse<String> response = Unirest.post("https://api.gravitex.ai/v1/images/generations")
  .header("Authorization", "<authorization>")
  .header("Content-Type", "<content-type>")
  .body("{\n  \"model\": \"<string>\",\n  \"prompt\": \"<string>\",\n  \"image\": [\n    \"<string>\"\n  ],\n  \"n\": 123,\n  \"size\": \"<string>\",\n  \"quality\": \"<string>\",\n  \"background\": \"<string>\",\n  \"output_format\": \"<string>\",\n  \"output_compression\": 123,\n  \"moderation\": \"<string>\",\n  \"user\": \"<string>\"\n}")
  .asString();

require 'uri'
require 'net/http'

url = URI("https://api.gravitex.ai/v1/images/generations")

http = Net::HTTP.new(url.host, url.port)
http.use_ssl = true

request = Net::HTTP::Post.new(url)
request["Authorization"] = '<authorization>'
request["Content-Type"] = '<content-type>'
request.body = "{\n  \"model\": \"<string>\",\n  \"prompt\": \"<string>\",\n  \"image\": [\n    \"<string>\"\n  ],\n  \"n\": 123,\n  \"size\": \"<string>\",\n  \"quality\": \"<string>\",\n  \"background\": \"<string>\",\n  \"output_format\": \"<string>\",\n  \"output_compression\": 123,\n  \"moderation\": \"<string>\",\n  \"user\": \"<string>\"\n}"

response = http.request(request)
puts response.read_body

POST

images

generations

GPT-Image-2

curl --request POST \
  --url https://api.gravitex.ai/v1/images/generations \
  --header 'Authorization: <authorization>' \
  --header 'Content-Type: <content-type>' \
  --data '
{
  "model": "<string>",
  "prompt": "<string>",
  "image": [
    "<string>"
  ],
  "n": 123,
  "size": "<string>",
  "quality": "<string>",
  "background": "<string>",
  "output_format": "<string>",
  "output_compression": 123,
  "moderation": "<string>",
  "user": "<string>"
}
'

import requests

url = "https://api.gravitex.ai/v1/images/generations"

payload = {
    "model": "<string>",
    "prompt": "<string>",
    "image": ["<string>"],
    "n": 123,
    "size": "<string>",
    "quality": "<string>",
    "background": "<string>",
    "output_format": "<string>",
    "output_compression": 123,
    "moderation": "<string>",
    "user": "<string>"
}
headers = {
    "Authorization": "<authorization>",
    "Content-Type": "<content-type>"
}

response = requests.post(url, json=payload, headers=headers)

print(response.text)

const options = {
  method: 'POST',
  headers: {Authorization: '<authorization>', 'Content-Type': '<content-type>'},
  body: JSON.stringify({
    model: '<string>',
    prompt: '<string>',
    image: ['<string>'],
    n: 123,
    size: '<string>',
    quality: '<string>',
    background: '<string>',
    output_format: '<string>',
    output_compression: 123,
    moderation: '<string>',
    user: '<string>'
  })
};

fetch('https://api.gravitex.ai/v1/images/generations', options)
  .then(res => res.json())
  .then(res => console.log(res))
  .catch(err => console.error(err));

<?php

$curl = curl_init();

curl_setopt_array($curl, [
  CURLOPT_URL => "https://api.gravitex.ai/v1/images/generations",
  CURLOPT_RETURNTRANSFER => true,
  CURLOPT_ENCODING => "",
  CURLOPT_MAXREDIRS => 10,
  CURLOPT_TIMEOUT => 30,
  CURLOPT_HTTP_VERSION => CURL_HTTP_VERSION_1_1,
  CURLOPT_CUSTOMREQUEST => "POST",
  CURLOPT_POSTFIELDS => json_encode([
    'model' => '<string>',
    'prompt' => '<string>',
    'image' => [
        '<string>'
    ],
    'n' => 123,
    'size' => '<string>',
    'quality' => '<string>',
    'background' => '<string>',
    'output_format' => '<string>',
    'output_compression' => 123,
    'moderation' => '<string>',
    'user' => '<string>'
  ]),
  CURLOPT_HTTPHEADER => [
    "Authorization: <authorization>",
    "Content-Type: <content-type>"
  ],
]);

$response = curl_exec($curl);
$err = curl_error($curl);

curl_close($curl);

if ($err) {
  echo "cURL Error #:" . $err;
} else {
  echo $response;
}

package main

import (
	"fmt"
	"strings"
	"net/http"
	"io"
)

func main() {

	url := "https://api.gravitex.ai/v1/images/generations"

	payload := strings.NewReader("{\n  \"model\": \"<string>\",\n  \"prompt\": \"<string>\",\n  \"image\": [\n    \"<string>\"\n  ],\n  \"n\": 123,\n  \"size\": \"<string>\",\n  \"quality\": \"<string>\",\n  \"background\": \"<string>\",\n  \"output_format\": \"<string>\",\n  \"output_compression\": 123,\n  \"moderation\": \"<string>\",\n  \"user\": \"<string>\"\n}")

	req, _ := http.NewRequest("POST", url, payload)

	req.Header.Add("Authorization", "<authorization>")
	req.Header.Add("Content-Type", "<content-type>")

	res, _ := http.DefaultClient.Do(req)

	defer res.Body.Close()
	body, _ := io.ReadAll(res.Body)

	fmt.Println(string(body))

}

HttpResponse<String> response = Unirest.post("https://api.gravitex.ai/v1/images/generations")
  .header("Authorization", "<authorization>")
  .header("Content-Type", "<content-type>")
  .body("{\n  \"model\": \"<string>\",\n  \"prompt\": \"<string>\",\n  \"image\": [\n    \"<string>\"\n  ],\n  \"n\": 123,\n  \"size\": \"<string>\",\n  \"quality\": \"<string>\",\n  \"background\": \"<string>\",\n  \"output_format\": \"<string>\",\n  \"output_compression\": 123,\n  \"moderation\": \"<string>\",\n  \"user\": \"<string>\"\n}")
  .asString();

require 'uri'
require 'net/http'

url = URI("https://api.gravitex.ai/v1/images/generations")

http = Net::HTTP.new(url.host, url.port)
http.use_ssl = true

request = Net::HTTP::Post.new(url)
request["Authorization"] = '<authorization>'
request["Content-Type"] = '<content-type>'
request.body = "{\n  \"model\": \"<string>\",\n  \"prompt\": \"<string>\",\n  \"image\": [\n    \"<string>\"\n  ],\n  \"n\": 123,\n  \"size\": \"<string>\",\n  \"quality\": \"<string>\",\n  \"background\": \"<string>\",\n  \"output_format\": \"<string>\",\n  \"output_compression\": 123,\n  \"moderation\": \"<string>\",\n  \"user\": \"<string>\"\n}"

response = http.request(request)
puts response.read_body

Introduction

GPT-Image-2 offers real-world intelligence, multilingual understanding, 4K resolution support and a smart routing layer, exposed through two complementary endpoints:

Endpoint	Focus	Description
`POST /v1/images/generations`	Text-to-image	Generate images straight from a text `prompt` — the recommended entry point for T2I
`POST /v1/images/edits`	Image-to-image	Professional editing: inpainting on the source image, multi-reference blending and other fine-grained edits; accepts both JSON and multipart/form-data

Both endpoints share the same response format: Base64 output (b64_json) plus usage token stats.

Capability comparison

Capability	OpenAI official API	Gravitex AI (generations endpoint)
Text-to-image	✅ Supported	✅ Supported
Image-to-image	❌ `edits` endpoint only	✅ Additionally supported
Mixed text + image input	❌	✅ Supported

In OpenAI’s official definition, /v1/images/generations for gpt-image-2 is a text-only endpoint: it accepts a text prompt and does not support passing images for image-to-image; official I2I is available only through /v1/images/edits. We have extended /v1/images/generations with image-to-image support — send a text prompt together with images and the model will re-create, restyle, reference or inpaint based on your input.

For full parameters and other image models, see Image generation.

Authentication

string

required

Bearer Token, e.g. Bearer sk-xxxxxxxxxx

string

required

application/json for JSON requests; multipart/form-data for file uploads

Supported models

Model ID	Description
`gpt-image-2`	OpenAI GPT-Image-2, multi-size including 4K, available on both endpoints

Parameters

Both endpoints accept the same parameter set.

string

required

Fixed value gpt-image-2

string

required

Image description (text-to-image) or edit instruction (image-to-image)

string | string[]

Input image(s) for image-to-image. A string for one image, an array for several; URL or base64 data URI. Omit for text-to-image

integer

default:"1"

Number of images to generate, 1-10

string

default:"1024x1024"

Image size, see the size table below

string

default:"high"

Image quality: low, medium, high, auto

string

default:"auto"

Background opacity: auto or opaque. Passing transparent returns an error

string

default:"png"

Output image format: png, jpeg

number

default:"100"

Compression level 0–100, applies to jpeg only

string

default:"auto"

Moderation strictness: auto or low (more permissive)

string

End-user identifier, used for abuse detection

size values

Value	Description
`1024x1024`	Square (default)
`1024x1536`	Portrait
`1536x1024`	Landscape
`2880x2880`	4K square
`2048x3072`	4K portrait
`3072x2048`	4K landscape
Custom `WxH`	Each dimension must be a multiple of 16; total pixels 655,360 – 8,294,400

quality values

Value	Description
`low`	Low quality, fastest generation
`medium`	Medium quality
`high`	High quality (default)
`auto`	Automatically pick the best quality

Text-to-image: `POST /v1/images/generations`

Generate an image from a text description. Pass model and prompt only, and the image is returned at the requested size and quality — ideal for creative generation and asset production driven purely by text.

Request examples

Basic
All parameters

curl -X POST "https://api.gravitex.ai/v1/images/generations" \
  -H "Authorization: Bearer sk-xxxxxxxxxx" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-image-2",
    "prompt": "A white cat running under the stars, cyberpunk style",
    "size": "1024x1536",
    "quality": "high",
    "n": 1
  }'

curl -X POST "https://api.gravitex.ai/v1/images/generations" \
  -H "Authorization: Bearer sk-xxxxxxxxxx" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-image-2",
    "prompt": "A beautiful medieval European princess with long golden hair and big blue eyes, wearing an ornate white lace gown and a crystal crown, standing on the balcony of an ancient castle in the sunlight.",
    "n": 1,
    "size": "2048x1152",
    "quality": "medium",
    "background": "opaque",
    "output_format": "jpeg",
    "output_compression": 95,
    "moderation": "low",
    "user": "biz_test_001"
  }'

Response example

{
  "created": 1786436787,
  "background": "opaque",
  "data": [
    {
      "b64_json": "iVBORw0KGgoAAAANSUh..."
    }
  ],
  "output_format": "png",
  "quality": "medium",
  "size": "2048x1152",
  "usage": {
    "input_tokens": 65,
    "input_tokens_details": {
      "image_tokens": 0,
      "text_tokens": 65
    },
    "output_tokens": 1413,
    "output_tokens_details": {
      "image_tokens": 1413,
      "text_tokens": 0
    },
    "total_tokens": 1478
  }
}

This endpoint also supports image-to-image. Add an image field to a text-to-image request (URL or base64 data URI — a string for one image, an array for several) and the request is routed to the image editing flow automatically: one unified endpoint, no switching required.There is no hard cap on the number of input images — the limit comes from the upstream total token budget. Each image may be up to 50MB.

Single image
Multi-image blend
base64 input

curl -X POST "https://api.gravitex.ai/v1/images/generations" \
  -H "Authorization: Bearer sk-xxxxxxxxxx" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-image-2",
    "prompt": "Add some tulips to this picture",
    "image": "https://example.com/input.png",
    "n": 1,
    "size": "2048x1152",
    "quality": "medium",
    "background": "opaque",
    "output_format": "jpeg",
    "output_compression": 95,
    "moderation": "low",
    "user": "biz_test_001"
  }'

curl -X POST "https://api.gravitex.ai/v1/images/generations" \
  -H "Authorization: Bearer sk-xxxxxxxxxx" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-image-2",
    "prompt": "Blend the style of the two images",
    "image": [
      "https://example.com/input1.jpeg",
      "https://example.com/input2.jpeg"
    ],
    "n": 1,
    "size": "2048x1152",
    "quality": "medium",
    "output_format": "jpeg",
    "output_compression": 95,
    "user": "biz_test_001"
  }'

curl -X POST "https://api.gravitex.ai/v1/images/generations" \
  -H "Authorization: Bearer sk-xxxxxxxxxx" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-image-2",
    "prompt": "Summer countryside scenery, clear blue sky with fluffy clouds, a winding clear stream, lush grass on both banks, 8K ultra HD, healing forest style",
    "image": "data:image/png;base64,/9j/4AAQSkZJRgABAQEASABIAAD/2...",
    "n": 1,
    "size": "2048x1152",
    "quality": "medium",
    "output_format": "jpeg",
    "output_compression": 95,
    "user": "biz_test_001"
  }'

Image-to-image responses use the same structure as text-to-image, except that usage.input_tokens_details.image_tokens is greater than 0, reflecting the tokens consumed by the input images:

{
  "created": 1786438823,
  "background": "opaque",
  "data": [
    {
      "b64_json": "iVBORw0KGgoAAAANSU..."
    }
  ],
  "output_format": "jpeg",
  "quality": "medium",
  "size": "2048x1152",
  "usage": {
    "input_tokens": 1048,
    "input_tokens_details": {
      "image_tokens": 1024,
      "text_tokens": 24
    },
    "output_tokens": 1413,
    "output_tokens_details": {
      "image_tokens": 1413,
      "text_tokens": 0
    },
    "total_tokens": 2461
  }
}

{
  "created": 1786439615,
  "background": "opaque",
  "data": [
    {
      "b64_json": "iVBORw0KGgoAAAANSUhEUgAACAAAAASACAIA..."
    }
  ],
  "output_format": "jpeg",
  "quality": "medium",
  "size": "2048x1152",
  "usage": {
    "input_tokens": 2064,
    "input_tokens_details": {
      "image_tokens": 2048,
      "text_tokens": 16
    },
    "output_tokens": 1413,
    "output_tokens_details": {
      "image_tokens": 1413,
      "text_tokens": 0
    },
    "total_tokens": 3477
  }
}

Image-to-image: `POST /v1/images/edits`

The standard OpenAI image editing endpoint: re-create, restyle, inpaint or blend multiple references on top of an input image. Accepts both JSON and multipart/form-data.

Option 1: multipart/form-data

Use this to upload local image files directly.

Images are passed via image[] — use image[] even for a single file.

Single image
Multiple images

curl -X POST "https://api.gravitex.ai/v1/images/edits" \
  -H "Authorization: Bearer sk-xxxxxxxxxx" \
  -F "model=gpt-image-2" \
  -F "prompt=Three-view reference sheet of a commercial fashion model, pure white background" \
  -F "image[]=@your-local-file.png" \
  -F "n=1" \
  -F "size=2048x1152" \
  -F "quality=medium" \
  -F "background=opaque" \
  -F "output_format=png" \
  -F "output_compression=100" \
  -F "moderation=low" \
  -F "user=biz_test_001"

curl -X POST "https://api.gravitex.ai/v1/images/edits" \
  -H "Authorization: Bearer sk-xxxxxxxxxx" \
  -F "model=gpt-image-2" \
  -F "prompt=Three-view reference sheet of a commercial fashion model, pure white background" \
  -F "image[]=@your-local-file-1.png" \
  -F "image[]=@your-local-file-2.png" \
  -F "n=1" \
  -F "size=2048x1152" \
  -F "quality=medium" \
  -F "background=opaque" \
  -F "output_format=png" \
  -F "output_compression=100" \
  -F "moderation=low" \
  -F "user=biz_test_001"

Option 2: JSON

Use this to pass image URLs or base64-encoded images. Parameters are identical to image-to-image on the generations endpoint.

Single image
Multiple images
base64 input

curl -X POST "https://api.gravitex.ai/v1/images/edits" \
  -H "Authorization: Bearer sk-xxxxxxxxxx" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-image-2",
    "prompt": "Add some tulips to this picture",
    "image": "https://example.com/input.png",
    "n": 1,
    "size": "2048x1152",
    "quality": "medium",
    "background": "opaque",
    "output_format": "jpeg",
    "output_compression": 95,
    "moderation": "low",
    "user": "biz_test_001"
  }'

curl -X POST "https://api.gravitex.ai/v1/images/edits" \
  -H "Authorization: Bearer sk-xxxxxxxxxx" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-image-2",
    "prompt": "Blend the style of the two images",
    "image": [
      "https://example.com/input1.png",
      "https://example.com/input2.jpeg"
    ],
    "n": 2,
    "size": "2048x1152",
    "quality": "medium",
    "background": "opaque",
    "output_format": "jpeg",
    "output_compression": 95,
    "moderation": "low",
    "user": "biz_test_001"
  }'

curl -X POST "https://api.gravitex.ai/v1/images/edits" \
  -H "Authorization: Bearer sk-xxxxxxxxxx" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-image-2",
    "prompt": "Change background to beach sunset",
    "image": "data:image/png;base64,/9j/4AAQSkZJRgABAQEASABIAAD/2...",
    "n": 1,
    "size": "1024x1024",
    "quality": "high"
  }'

This endpoint also supports text-to-image. Omit the image field and send only model and prompt: the request is handled as pure text-to-image and returns the same result as /v1/images/generations, so apps already wired to edits can keep a single endpoint.

curl -X POST "https://api.gravitex.ai/v1/images/edits" \
  -H "Authorization: Bearer sk-xxxxxxxxxx" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-image-2",
    "prompt": "A white cat running under the stars, cyberpunk style",
    "size": "1024x1536",
    "quality": "high",
    "n": 1
  }'

Response fields

Both endpoints return the same structure:

Field	Type	Description
`created`	integer	Creation timestamp
`background`	string	Background type (`opaque` / `auto`)
`data`	array	Generated images
`data[].b64_json`	string	Base64-encoded image data
`output_format`	string	Output format (`png` / `jpeg`)
`quality`	string	Quality actually used
`size`	string	Size actually used
`usage`	object	Token usage stats

usage fields

Field	Description
`input_tokens`	Total input tokens (text + image)
`output_tokens`	Total output tokens (image output)
`total_tokens`	Total tokens
`input_tokens_details.text_tokens`	Text input tokens
`input_tokens_details.image_tokens`	Image input tokens (0 for T2I, > 0 for I2I)

GPT-Image-2 always returns base64-encoded image data (b64_json); response_format=url is not supported.
output_tokens is entirely image output — this model produces no text output.
Billing is token-based, with separate unit prices for text input, image input and image output.

Pricing

All prices are in USD per 1M tokens, billed on the usage tokens returned upstream.

Type	Price ($/1M tokens)
Text Input Tokens	$5.00
Image Input Tokens	$8.00
Cached Text Input Tokens	$1.25
Cached Image Input Tokens	$2.00
Image Output Tokens	$30.00

Notes

Image generation usually takes 10–30 seconds, depending on size and quality
Image data is always returned base64-encoded
Supported input formats: PNG, JPEG, WebP; output formats: PNG, JPEG
4K generation takes noticeably longer — prefer standard sizes when possible
Setting quality to low speeds up generation significantly
gpt-image-2 always processes input images at high fidelity; the input_fidelity parameter is neither needed nor supported
Each uploaded image may be up to 50MB
n (number of output images) ranges from 1 to 10
There is no hard cap on input image count — it is bounded by the upstream total token limit
The model has no text output; output_tokens is entirely image output tokens
background supports only auto (default) and opaque — transparent returns an error

Image generation

Overview of the multi-model image generation API

Image edits

More usage and examples for the edits endpoint

API documentation

Chat & text

Safety & audio

Image Series

Video Series

Completions & Embeddings

Interface module

Introduction

Capability comparison

Authentication

Supported models

Parameters

size values

quality values

Text-to-image: `POST /v1/images/generations`

Request examples

Response example

Image-to-image: `POST /v1/images/edits`

Option 1: multipart/form-data

Option 2: JSON

Response fields

usage fields

Pricing

Notes

Image generation

Image edits

​Introduction

​Capability comparison

​Authentication

​Supported models

​Parameters

​size values

​quality values

​Text-to-image: POST /v1/images/generations

​Request examples

​Response example

​Image-to-image: POST /v1/images/edits

​Option 1: multipart/form-data

​Option 2: JSON

​Response fields

​usage fields

​Pricing

​Notes

​Related resources

Image generation

Image edits

Introduction

Capability comparison

Authentication

Supported models

Parameters

size values

quality values

Text-to-image: `POST /v1/images/generations`

Request examples

Response example

Image-to-image: `POST /v1/images/edits`

Option 1: multipart/form-data

Option 2: JSON

Response fields

usage fields

Pricing

Notes

Related resources