Create an image

Depending on the model, you can control the number of images, size, quality, style, and response format.

POST/v1/images/generations

Generate one or more images from a text prompt. Describe the subject, scene, composition, viewpoint, visual style, and overall mood, then use the size, quality, and response-format options supported by the model.

Model availability and constraints for fields such as n, quality, size, and style depend on the selected model. Use an image-generation model visible to the current API key.

Authorization

bearerAuth

AuthorizationBearer <token>

Use Bearer your-api-key in the Authorization Header.

In: header

Request Body

application/json

Image generation request parameters.

TypeScript Definitions

Use the request body type in TypeScript.

model?string

The image generation model ID to be used. Please select models that are visible to the current API Key and support image generation; Different models may support different parameters and limitations.

Length1 <= length

prompt*string

Text description of the image you expect to generate. It is recommended to clarify the subject, key visual elements, composition, perspective, artistic style, and atmosphere

DALL· E 2 accepts up to 1,000 characters, DALL· E 3 accepts up to 4,000 characters; Other models are subject to limitations based on their actual capabilities. If generation fails, please also check whether the prompt complies with the content policy

Length1 <= length

n?integer|null

The number of images to be generated, with values ranging from 1 to 10. DALL· E 3 only supports n=1; whether other models support multi-image return depends on their actual capabilities.

Range1 <= value <= 10

Default1

quality?string

Image quality. hd Can generate more detailed and consistent images, and is only applicable to DALL· E 3; Other models may use different values.

Default"standard"

Value in

"standard"
"hd"

response_format?string|null

Image response format. url returns a temporary image URL that remains valid for 60 minutes after generation; b64_json returns Base64-encoded image data.

Default"url"

Value in

"url"
"b64_json"
null

size?string|null

Image size. DALL· E 2 supports 256x256, 512x512, 1024x1024, DALL· E 3 supports 1024x1024, 1792x1024, 1024x1792.

Default"1024x1024"

Value in

"256x256"
"512x512"
"1024x1024"
"1792x1024"
"1024x1792"
null

style?string|null

Image style. vivid Leaning towards surreal and dramatic effects, natural Leaning toward natural and realistic expression. This specification applies only to DALL· E 3。

Default"vivid"

Value in

"vivid"
"natural"
null

user?string

Unique identifier for the end user, used to help monitor and detect abuse.

Length1 <= length

Response Body

`application/json`

curl -X POST "https://example.com/v1/images/generations" \  -H "Content-Type: application/json" \  -d '{    "model": "MODEL_ID",    "prompt": "A cute little otter playing in the clear water, with soft natural light and clear details",    "n": 1,    "size": "1024x1024"  }'

{
  "created": 1589478378,
  "data": [
    {
      "url": "https://example.com/generated-image.png",
      "revised_prompt": "A cute little otter playing in the clear water, with round eyes and fluffy fur."
    }
  ]
}

Authorization

Request Body

Response Body

200application/json

400application/json

401application/json

403application/json

`application/json`

`application/json`

`application/json`

`application/json`