Create Image

Creates an image given a prompt using a GPT Image model.

Endpoint

POST /v1/images/generations

Authentication

Type	Description
Bearer Token	Include Authorization: Bearer <API_key> in the request header for account verification.

Request Headers

Name	Type	Required	Description
Content-Type	string	Yes	Set request media type to application/json
Authorization	string	Yes	Bearer API_key, used for account verification

Request Body

Name	Type	Required	Default	Description
model	string	Yes	-	The GPT image model to use for image generation; must be specified explicitly. See "Supported Model IDs" below
prompt	string	Yes	-	A text description of the desired image(s), up to 32000 characters
background	string	No	auto	Background of the generated image(s): transparent, opaque, or auto. See Parameter Details below
moderation	string	No	auto	Content-moderation level: low (less restrictive) or auto. Only supported for GPT image models
n	integer	No	1	Number of images to generate, between 1 and 10
output_compression	integer	No	100	Compression level (0-100%), only for GPT image models with webp or jpeg output formats
output_format	string	No	png	Format of the returned images: png, jpeg, or webp. Only supported for GPT image models
partial_images	integer	No	0	Number of partial images in streaming responses, 0-3; when 0, a single image is sent in one streaming event
quality	string	No	auto	Quality of the generated image. See Parameter Details below
response_format	string	No	-	Legacy parameter for retired image models. Unsupported for GPT image models (always return base64-encoded images)
safety_identifier	string	No	-	A stable end-user identifier for safety monitoring, max 128 characters. See Parameter Details below
size	string	No	auto	Size of the generated images. See Parameter Details below
stream	boolean	No	false	Whether to generate the image in streaming mode. Only supported for GPT image models
style	string	No	-	Legacy parameter for retired image models. Unsupported for GPT image models; describe the desired style in the prompt instead
user	string	No	-	Deprecated: use safety_identifier instead. Used when safety_identifier is omitted, null, or empty

Supported Model IDs:

• gp-im-1.5
• gp-im-2
• gp-im-2-2026-04-21
• gp-im-2.5-su
• gp-im-2.5-su-2026-09-08
• gp-im-2.5-fl
• gp-im-2.5-fl-2026-09-08
• gp-im-1-mini
• gp-im-1 (shut down on October 23, 2026; use gp-im-2.5-su or gp-im-2.5-fl instead)
• dall-e-2
• dall-e-3

Parameter Details

background

This parameter is only supported for the GPT image models. When auto is used, the model will automatically determine the best background for the image.

gp-im-2.5-su and gp-im-2.5-fl, including their 2026-09-08 snapshots, support opaque and transparent backgrounds. For gp-im-2 and gp-im-2-2026-04-21, transparent background support is in preview. When using transparent, set output_format to png or webp.

quality Enum Values

• auto: automatically selects the best quality for the given model (default)
• low / medium / high: supported for the GPT image models
• xhigh / max: only supported for gp-im-2.5-su and gp-im-2.5-fl, including their 2026-09-08 snapshots
• standard / hd: legacy enum values for retired image models

size

For gp-im-2, gp-im-2-2026-04-21, gp-im-2.5-su, gp-im-2.5-su-2026-09-08, gp-im-2.5-fl, and gp-im-2.5-fl-2026-09-08, arbitrary resolutions are supported as WIDTHxHEIGHT strings, for example 1536x864. Width and height must both be divisible by 16 and the requested aspect ratio must be between 1:3 and 3:1. Resolutions above 2560x1440 are experimental, and the maximum supported resolution is 3840x2160. The requested size must also satisfy the model's current pixel and edge limits.

Standard sizes: 1024x1024, 1536x1024, 1024x1536; auto is supported for models that allow automatic sizing. Legacy enum values: 256x256, 512x512, 1792x1024, 1024x1792.

safety_identifier

A stable identifier for an end user in your organization, used for safety monitoring and abuse detection. Hash a username or email address to avoid sending identifying information. A non-empty value takes precedence over the deprecated user parameter. If omitted, null, or empty, the API uses user when available.

Example Request

Generate image

curl --request POST \
  --url https://neolink.com/api/v1/images/generations \
  --header 'Authorization: Bearer <API Key>' \
  --header 'Content-Type: application/json' \
  --data '{
    "model": "gp-im-2",
    "prompt": "A cute baby sea otter",
    "n": 1,
    "size": "1024x1024"
  }'

Streaming

curl --request POST \
  --url https://neolink.com/api/v1/images/generations \
  --header 'Authorization: Bearer <API Key>' \
  --header 'Content-Type: application/json' \
  --data '{
    "model": "gp-im-2",
    "prompt": "A cute baby sea otter",
    "n": 1,
    "size": "1024x1024",
    "stream": true
  }'

Full parameters

curl --request POST \
  --url https://neolink.com/api/v1/images/generations \
  --header 'Authorization: Bearer <API Key>' \
  --header 'Content-Type: application/json' \
  --data '{
    "model": "gp-im-2.5-fl",
    "prompt": "A cute baby sea otter",
    "background": "transparent",
    "moderation": "low",
    "n": 1,
    "output_compression": 100,
    "output_format": "png",
    "partial_images": 1,
    "quality": "medium",
    "safety_identifier": "user-1234",
    "size": "1024x1024"
  }'

Response Body

Name	Type	Description
created	integer	The Unix timestamp (in seconds) of when the image was created
background	string	The background parameter used: transparent or opaque
data	array	The list of generated images
data.b64_json	string	The base64-encoded JSON of the generated image, returned by default for GPT image models
data.revised_prompt	string	The revised prompt used to generate the image, for models that support prompt revision. Not returned by GPT image models
data.url	string	The URL of the generated image (returned when the legacy response_format is url). Unsupported for GPT image models
output_format	string	The output format: png, webp, or jpeg
quality	string	The quality of the image: low, medium, high, xhigh, or max
safety_identifier	string	The effective safety identifier: the non-empty safety_identifier from the request, or the deprecated user value when it is omitted, null, or empty
size	string	The image dimensions as a WIDTHxHEIGHT string, for example 1536x864
usage	object	Token usage information, for gp-im-1 only
usage.input_tokens	integer	The number of tokens (images and text) in the input prompt
usage.input_tokens_details.image_tokens	integer	The number of image tokens in the input prompt
usage.input_tokens_details.text_tokens	integer	The number of text tokens in the input prompt
usage.output_tokens	integer	The number of output tokens generated by the model
usage.total_tokens	integer	The total number of tokens (images and text) used for the image generation
usage.output_tokens_details.image_tokens	integer	The number of image output tokens generated by the model
usage.output_tokens_details.text_tokens	integer	The number of text output tokens generated by the model

Response Example

{
  "created": 1713833628,
  "data": [
    {
      "b64_json": "..."
    }
  ],
  "usage": {
    "total_tokens": 100,
    "input_tokens": 50,
    "output_tokens": 50,
    "input_tokens_details": {
      "text_tokens": 10,
      "image_tokens": 40
    }
  }
}

Streaming Response

When stream: true is set, the response is returned as an SSE event stream with the following events:

• image_generation.partial_image: a partial image, with partial_image_index indicating its index; the final image may be sent before the full number of partial_images is reached if it is generated more quickly
• image_generation.completed: the final full image with token usage

event: image_generation.partial_image
data: {"type":"image_generation.partial_image","b64_json":"...","partial_image_index":0}

event: image_generation.completed
data: {"type":"image_generation.completed","b64_json":"...","usage":{"total_tokens":100,"input_tokens":50,"output_tokens":50,"input_tokens_details":{"text_tokens":10,"image_tokens":40}}}