GPT Image 2 API — OpenAI's Next-Gen Image Model

Generate images with GPT Image 2 (OpenAI's next-gen image model) through VicSee API. Up to 4K resolution, 11 aspect ratios, image-to-image editing. 8-20 credits per image.

GPT Image 2 is OpenAI's next-generation image model — strong photorealism, sharper text rendering, and polished design output. Available on VicSee for both text-to-image and image-to-image (editing) workflows.

Pricing

Resolution	Credits	Price (Pro Yearly)	Price (Pro Monthly)
1K	8	$0.048	$0.096
2K	12	$0.072	$0.144
4K	20	$0.120	$0.240

Credits are deducted only on successful generation. Failed tasks are auto-refunded.

Endpoint

POST https://vicsee.com/api/v1/generate

See Authentication for API key setup.

Text to Image

Generate images from a text prompt.

Request Parameters

Parameter	Type	Required	Description
model	string	Yes	`gpt-image-2-text-to-image`
input.prompt	string	Yes	Description of the image (up to 20,000 characters)
input.resolution	string	No	`1K`, `2K`, or `4K` (default: `1K`)
input.aspect_ratio	string	No	See aspect ratios (default: `auto`)

Example Request

curl -X POST https://vicsee.com/api/v1/generate \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-image-2-text-to-image",
    "input": {
      "prompt": "A studio product shot of a ceramic coffee mug on a marble countertop, soft natural light, photorealistic",
      "resolution": "2K",
      "aspect_ratio": "1:1"
    }
  }'

Image to Image

Edit or transform a reference image with a text prompt.

Request Parameters

Parameter	Type	Required	Description
model	string	Yes	`gpt-image-2-image-to-image`
input.prompt	string	Yes	Description of the edit or transformation
input.image_urls	array	Yes	Reference image URLs (jpeg/png/webp, max 30MB each)
input.resolution	string	No	`1K`, `2K`, or `4K` (default: `1K`)
input.aspect_ratio	string	No	Output aspect ratio (default: `auto`)

Example Request

curl -X POST https://vicsee.com/api/v1/generate \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-image-2-image-to-image",
    "input": {
      "prompt": "Place this product on a wooden desk with warm lighting and a coffee cup beside it",
      "image_urls": ["https://example.com/product.jpg"],
      "resolution": "2K",
      "aspect_ratio": "16:9"
    }
  }'

Supported Aspect Ratios

GPT Image 2 supports 11 aspect ratios:

Ratio	Type	Use Case
`auto`	Automatic	Model selects best ratio (1K only)
`1:1`	Square	Social media, profile pictures
`5:4`	Landscape	Desktop wallpapers
`9:16`	Vertical	Stories, Reels, TikTok
`21:9`	Ultrawide	Cinematic banners
`16:9`	Widescreen	YouTube thumbnails, presentations
`4:3`	Standard	Blog images, traditional photos
`3:2`	Landscape	Photography, prints
`4:5`	Social	Instagram posts
`3:4`	Portrait	Portraits, product photos
`2:3`	Portrait	Posters, book covers

Resolution Constraints

auto aspect ratio outputs 1K only. Specify a fixed ratio (e.g. 16:9) to generate at 2K or 4K.
1:1 aspect ratio cannot output 4K. Use 1K or 2K for square outputs.

Requests that violate these constraints are rejected by the model and credits are auto-refunded.

Response

See Tasks for polling task status and retrieving results.

{
  "task_id": "abc123",
  "status": "pending"
}

For error handling, see Errors.

GPT Image 2

On this page