Generate Images Using GPT1.5

Sample Request

curl --location 'https://cloud.leonardo.ai/api/rest/v2/generations' \
     --header 'authorization: Bearer {api-key}' \
     --header 'Content-Type: application/json' \
       --data '
{
  "public": false,
  "model": "gpt-image-1.5",
  "parameters": {
    "mode": "QUALITY",
    "prompt": "Koala with purple hat",
    "quantity": 2,
    "width": 1024,
    "height": 1024,
    "seed": 4294967295,
    "guidances": {
      "image_reference": [
        {
          "image": {
            "id": "00000000-0000-0000-0000-000000000001",
            "type": "UPLOADED"
          },
          "strength": 0.5,
          "prompt_enhance": "OFF"
        }
      ]
    }
  }
}
'

Parameters

model - Set to gpt-image-1.5
height - Supports 1024 and 1536
width - Supports 1024 and 1536
quantity - Number of images generated
mode - FAST, QUALITY, ULTRA
guidances.image_reference - List of image references up to 6. Supports uploaded images and images generated on the Leonardo platform.
seed - Randomisation parameter. The maximum seed number is 4294967295

Aspect Ratio Settings

Aspect Ratio	Width	Height
2:3	1024	1536
1:1	1024	1024
3:2	1536	1024

Cost

Pricing is a based on the input height, width, quantity, mode and whether or not image references were used. Refer to the calculation below to get the API credit cost based on your settings.

Base cost = 1.23977e-5

ImageRef = 1.2 if request has image guidance

Quality = 2.92 for Medium or 10.538 for High

Cost = Quantity * BaseCost * Height*Width * Quality * ImageRef

	Multiplier	Notes	Cost
FAST	1	13 token for 1024x1024 t2i	13
QUALITY	2.92	Based on 1024x1024 output	38
ULTRA	10.538	Based on 1024x1024 output	137
Quantity	Self
Height*Width	Self
Image References	1.2	flat 20% increase regardless of number input images