GPT Image 2

client.images.generate(params)

POST/v1/images/generations

Generate images with the provider-compatible OpenAI Images API. This synchronous endpoint returns image data directly.

Request body

Use the native OpenAI Images request shape. The model must be gpt-image-2.

ImageGenerateParams

stringmodelrequired

Use gpt-image-2 on this endpoint.

Default: gpt-image-2

stringpromptrequired

Text description of the image to generate.

Optional<integer>n

Number of images to generate when supported.

Optional<string>size

Output dimensions supported by the model.

Optional<string>quality

Output quality such as low

Optional<string>background

Background setting such as transparent

Optional<string>output_format

Output encoding such as png

Optional<integer>output_compression

Output compression level from 0 to 100.

Optional<string>moderation

Provider-compatible moderation setting.

Optional<string>user

Provider-compatible end-user identifier.

curl https://api.sandbase.ai/v1/images/generations \
  -H "Authorization: Bearer $SANDBASE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"gpt-image-2","prompt":"A paper-cut city floating above the clouds","size":"1024x1024"}'

Response200 OK

{
  "created": 1787529600,
  "data": [
    {
      "b64_json": "iVBORw0KGgo..."
    }
  ],
  "usage": {
    "input_tokens": 12,
    "output_tokens": 4096,
    "total_tokens": 4108
  }
}

Native protocol notes

This model uses the OpenAI Images API contract, not the general SandBase /v1/run model endpoint. Requests are synchronous and return the generated image in data as b64_json or a URL.

Use the shared image generation reference for authentication, response details, limits, and error handling. The SandBase catalog entry is openai/gpt-image-2-official; because this endpoint preserves the OpenAI Images protocol, send the provider model name gpt-image-2 in the request body.

Request body

Native protocol notes ​

Official OpenAI resources ​

Native protocol notes

Official OpenAI resources