> ## Documentation Index
> Fetch the complete documentation index at: https://docs.nano-gpt.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Image Models

> List available image generation and image editing models with capabilities and supported parameters

## Overview

Use `GET /api/v1/image-models` to discover the currently available image models and their capabilities. Do not hardcode image model capability tables in your client; model availability and supported parameters can change.

For new integrations using the normalized Image API, prefer [`GET /api/v1/images/models`](/api-reference/endpoint/image-api-models). That newer route returns model-specific `supported_parameters`, endpoint metadata links, and input-reference capabilities for use with [`POST /api/v1/images`](/api-reference/endpoint/image-api-generate).

This endpoint is cacheable. Refresh it periodically and handle new fields as additive.

## Endpoint

```text theme={null}
GET https://nano-gpt.com/api/v1/image-models
```

## Authentication

Authentication is optional.

* `Authorization: Bearer YOUR_API_KEY`
* `x-api-key: YOUR_API_KEY`

## Query Parameters

| Parameter  | Type    | Default | Description                                                                   |
| ---------- | ------- | ------- | ----------------------------------------------------------------------------- |
| `detailed` | boolean | `true`  | Include names, descriptions, pricing, capabilities, and supported parameters. |

## Response

```json theme={null}
{
  "object": "list",
  "data": [
    {
      "id": "pruna-ai/p-image/text-to-image",
      "object": "model",
      "created": 1778544000,
      "owned_by": "prunaai",
      "name": "P-Image",
      "description": "Fast text-to-image generation for low-cost drafts, social visuals, and rapid creative iteration.",
      "architecture": {
        "modality": "text->image",
        "input_modalities": ["text"],
        "output_modalities": ["image"]
      },
      "pricing": {
        "per_image": {
          "1024*1024": 0.005,
          "1376*768": 0.005
        },
        "currency": "USD"
      },
      "capabilities": {
        "image_generation": true,
        "image_to_image": false,
        "inpainting": false,
        "nsfw": false
      },
      "supported_parameters": {
        "resolutions": ["1024x1024", "1376x768", "1184x896"],
        "max_images": 4
      },
      "icon_url": "/icons/PrunaAI.svg",
      "tags": ["text-to-image", "prunaai", "wavespeed"],
      "category": "image"
    }
  ],
  "meta": {
    "count": 169,
    "generated_at": "2026-05-14T00:00:00.000Z"
  }
}
```

## Supported Parameters

Supported parameters vary by model. Common fields include:

* `resolutions`: array of supported resolution values for this model.
* `max_images`: maximum number of images that can be requested at once.
* `rendering_speed`: available rendering speed values, when the model exposes speed tiers.
* `fixed_image_count`: fixed output count for models that do not allow custom `n`.

Use `supported_parameters.resolutions` to populate resolution or size controls. Treat the values as model-specific strings. Most image generation models use pixel dimensions such as `1024x1024`, but some models use values such as `auto`, aspect ratios, quality tiers, or scale factors.

Use the model's `supported_parameters` and `capabilities` objects to decide which controls to show and which request fields to send.

## Fields

| Field                                    | Type      | Description                                                                                                                                                               |
| ---------------------------------------- | --------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| `id`                                     | string    | Model ID to pass to image generation or image editing endpoints.                                                                                                          |
| `object`                                 | string    | Always `model`.                                                                                                                                                           |
| `created`                                | integer   | Unix timestamp derived from the model's added date.                                                                                                                       |
| `owned_by`                               | string    | Provider or owner identifier.                                                                                                                                             |
| `name`                                   | string    | Human-readable model name.                                                                                                                                                |
| `description`                            | string    | Short model description.                                                                                                                                                  |
| `architecture`                           | object    | Modality metadata, including input and output modalities.                                                                                                                 |
| `pricing`                                | object    | USD pricing metadata. Shape can vary by model.                                                                                                                            |
| `capabilities`                           | object    | Feature flags such as `image_generation`, `image_to_image`, and `inpainting`.                                                                                             |
| `capabilities.nsfw`                      | boolean   | Whether the model is uncensored / NSFW-capable.                                                                                                                           |
| `supported_parameters`                   | object    | Model-specific parameters such as sizes, counts, quality settings, seeds, or image input controls.                                                                        |
| `supported_parameters.resolutions`       | string\[] | Model-specific supported output resolution values. Usually pixel dimensions like `1024x1024`; may also be symbolic values such as `auto` or model-specific scale factors. |
| `supported_parameters.max_images`        | integer   | Maximum number of images the model supports per request.                                                                                                                  |
| `supported_parameters.rendering_speed`   | string\[] | Optional list of rendering speed tiers. Present only for models that expose speed choices.                                                                                |
| `supported_parameters.fixed_image_count` | integer   | Optional fixed output image count. Present when the model always returns a fixed number of images.                                                                        |
| `icon_url`                               | string    | Optional provider icon path.                                                                                                                                              |
| `tags`                                   | string\[] | Optional model tags.                                                                                                                                                      |
| `category`                               | string    | Model category. Usually `image`.                                                                                                                                          |

`supported_parameters`, including `supported_parameters.resolutions`, is included only in detailed responses. `detailed=true` is the default. If you pass `detailed=false`, each item may include only basic identifiers such as `id`, `object`, `created`, and `owned_by`.

## Example

```bash theme={null}
curl "https://nano-gpt.com/api/v1/image-models?detailed=true" \
  -H "Authorization: Bearer $NANOGPT_API_KEY"
```

## Notes

* Supported parameters vary by model.
* Use this endpoint instead of hardcoding media model capabilities.
* The response is cacheable, but model availability can change.
* Pricing values are in USD unless otherwise documented in the response.