Generic Image Generation

Use /v1/images/generations for nano-banana families and gpt-image-2 image models.

/v1/images/generations is the recommended generic image entrypoint in ModelSell. For text-to-image, send model and prompt. For reference-image generation or image editing, also send images, image, or image_urls.

Supported parameters are not identical across models. Keep the unified entrypoint, but choose fields and values according to the requested model.

Endpoints

Operation	Endpoint
Image generation	`POST /v1/images/generations`
Image edit compatible entrypoint	`POST /v1/images/edits`
Poll async image task	`GET /v1/images/generations/{task_id}`

Supported Models

Model	Description
`nano-banana`	Base nano-banana image generation and editing model.
`nano-banana-2`	nano-banana 2 image generation and editing model with broader resolution and search-enhancement options.
`nano-banana-2-lite`	Lightweight nano-banana 2 model for faster or lower-cost image generation and editing.
`nanp-banana-pro`	nano-banana pro image generation and editing model.
`gpt-image-2`	GPT Image 2 image generation and editing model.

Generic Request Fields

Field	Type	Required	Description
`model`	string	Yes	Model name, such as `nano-banana-2` or `gpt-image-2`.
`prompt`	string	Yes	Image generation or edit prompt.
`images`	string[]	No	Reference image URLs or reachable image references. This is the recommended unified field.
`image`	string / string[]	No	OpenAI image edit compatible field for single or multiple image inputs.
`image_urls`	string[]	No	Reference image field compatible with some upstream formats.
`aspect_ratio`	string	No	Output aspect ratio. Suitable for nano-banana families and some gpt-image-2 channels.
`resolution`	string	No	Image resolution or billing tier. Suitable for nano-banana families and some gpt-image-2 channels.
`size`	string	No	OpenAI/gpt-image compatible size field. Suitable for `gpt-image-2`.
`quality`	string	No	Image quality. Suitable for `gpt-image-2`; some channels may allow it for other models.
`output_format`	string	No	Output format, such as `png` or `jpeg`.
`response_format`	string	No	OpenAI-compatible response format, such as `url` or `b64_json`.
`n`	number	No	Number of images. Support depends on the model and channel.
`enable_web_search`	boolean	No	Available for nano-banana-2 families to enable web search enhancement.
`enable_image_search`	boolean	No	Available for nano-banana-2 families to enable image search enhancement.

Model-Specific Parameters

nano-banana Families

The nano-banana family includes nano-banana, nano-banana-2, nano-banana-2-lite, and nanp-banana-pro. Prefer images, aspect_ratio, resolution, and output_format.

Field	Recommended values
`images`	Reference image array. Omit it for text-to-image; send it for image-to-image or image editing.
`aspect_ratio`	`1:1`, `3:2`, `2:3`, `3:4`, `4:3`, `4:5`, `5:4`, `9:16`, `16:9`, `21:9`, `1:4`, `4:1`, `1:8`, `8:1`.
`resolution`	`0.5k`, `1k`, `2k`, `4k`.
`output_format`	`png`, `jpeg`.
`enable_web_search`	`true` or `false`, mainly for nano-banana-2 families.
`enable_image_search`	`true` or `false`, mainly for nano-banana-2 families.

{
  "model": "nano-banana-2",
  "prompt": "Replace the item in the poster with a cup",
  "images": [
    "https://example.com/input.png"
  ],
  "aspect_ratio": "4:3",
  "resolution": "2k",
  "output_format": "png",
  "enable_web_search": false,
  "enable_image_search": false
}

gpt-image-2

gpt-image-2 may use either OpenAI-compatible size parameters or tiered image parameters, depending on the active channel. Do not treat size and the nano-banana resolution tier as the same semantic field.

Field	Recommended values
`size`	OpenAI-compatible sizes such as `1024x1024`, `1536x1024`, or `1024x1536`, depending on channel support.
`aspect_ratio`	Ratio-based channels may use values such as `1:1`, `4:3`, `16:9`, or `9:16`.
`resolution`	Tier-based channels may use values such as `1k`, `2k`, or `4k`.
`quality`	`low`, `medium`, `high`, `auto`.
`output_format`	`png`, `jpeg`.
`response_format`	`url`, `b64_json`.

{
  "model": "gpt-image-2",
  "prompt": "A cinematic product photo of a ceramic cup",
  "size": "1536x1024",
  "quality": "high",
  "response_format": "url"
}

If the active channel uses ratio and tier fields, call it like this:

{
  "model": "gpt-image-2",
  "prompt": "A cinematic product photo of a ceramic cup",
  "aspect_ratio": "4:3",
  "resolution": "2k",
  "quality": "high",
  "output_format": "png"
}

Image Editing

For reference-image input, prefer the images array. image also works for single-image input, but images is clearer for multi-image editing.

{
  "model": "nano-banana-2",
  "prompt": "Replace the main object in the reference poster with a cup",
  "images": [
    "https://example.com/poster.png"
  ],
  "aspect_ratio": "4:3",
  "resolution": "2k"
}

Response

Some image models return an async task ID:

{
  "id": "task_xxx",
  "object": "image.task",
  "created": 1770000000,
  "model": "nano-banana-2",
  "status": "created"
}

After the task completes, poll the async task endpoint for the final image URL. Synchronous waiting and OpenAI image response conversion depend on the active model and channel configuration.