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

# Image models

> Every image model available through the Deepshi API, with sizing options, prompt limits, and capabilities.

Use these models with the [image generation](/capabilities/image-generation) and [image editing](/capabilities/image-editing) endpoints. Reference a model by its **id** in the `model` field.

| Sizing style                                                       | What to send                                                   |
| ------------------------------------------------------------------ | -------------------------------------------------------------- |
| <Icon icon="ruler-combined" /> Pixel dimensions                    | `width` and `height`                                           |
| <Icon icon="up-right-and-down-left-from-center" /> Resolution tier | An `aspect_ratio` plus a `resolution` like `1K`, `2K`, or `4K` |
| <Icon icon="crop" /> Aspect ratio                                  | Just an `aspect_ratio`                                         |
| <Icon icon="image" /> Preset size                                  | A named `image_size`                                           |

<Note>
  The exact set your key can call, and each model's supported parameters, is returned by
  `GET /v1/models`. For per-model rates, see [deepshi.ai](https://deepshi.ai/).
</Note>

## Deepshi models

Deepshi's own image models, running on Deepshi GPUs.

<CardGroup cols={2}>
  <Card title="Deepshi Banana Pro" icon="bolt">
    `deepshi-banana-pro`

    Fast in-house text-to-image, tuned for quick, clean generations at any size.

    **Prompt** 10K · <Tooltip tip="Pixel dimensions"><Icon icon="ruler-combined" /></Tooltip>
  </Card>

  <Card title="Deepshi Image 1" icon="layer-group">
    `deepshi-image-1`

    In-house model for text-to-image and multi-reference editing.

    **Prompt** 10K · <Tooltip tip="Pixel dimensions"><Icon icon="ruler-combined" /></Tooltip> <Tooltip tip="Image editing"><Icon icon="pen" /></Tooltip>
  </Card>
</CardGroup>

## Third-party models

Leading image models from across the industry, through the same endpoint, key, and balance.

<CardGroup cols={2}>
  <Card title="Nano Banana 2">
    `nano-banana-2`

    Google's fast, prompt-faithful generator with flexible aspect ratios, up to 4K.

    **Prompt** 50K · <Tooltip tip="Resolution tier"><Icon icon="up-right-and-down-left-from-center" /></Tooltip> <Tooltip tip="Image editing"><Icon icon="pen" /></Tooltip>
  </Card>

  <Card title="Nano Banana Pro">
    `nano-banana-pro`

    The pro tier of Nano Banana, for higher-fidelity detail and large 4K renders.

    **Prompt** 50K · <Tooltip tip="Resolution tier"><Icon icon="up-right-and-down-left-from-center" /></Tooltip> <Tooltip tip="Image editing"><Icon icon="pen" /></Tooltip>
  </Card>

  <Card title="GPT Image 1.5">
    `gpt-image-1.5`

    OpenAI's image model, with strong prompt adherence and text rendering.

    **Prompt** 32K · <Tooltip tip="Preset size"><Icon icon="image" /></Tooltip> <Tooltip tip="Image editing"><Icon icon="pen" /></Tooltip>
  </Card>

  <Card title="GPT Image 2">
    `gpt-image-2`

    The latest GPT Image model, with sharper detail and a selectable quality tier.

    **Prompt** 32K · <Tooltip tip="Preset size"><Icon icon="image" /></Tooltip> <Tooltip tip="Image editing"><Icon icon="pen" /></Tooltip>
  </Card>

  <Card title="Seedream 4.5">
    `seedream-4.5`

    ByteDance's photorealistic generator with flexible preset sizes up to 4K.

    **Prompt** 10K · <Tooltip tip="Preset size"><Icon icon="image" /></Tooltip> <Tooltip tip="Image editing"><Icon icon="pen" /></Tooltip>
  </Card>

  <Card title="Seedream 5 Lite">
    `seedream-5-lite`

    A lighter, faster Seedream variant for quick high-resolution generations.

    **Prompt** 10K · <Tooltip tip="Preset size"><Icon icon="image" /></Tooltip> <Tooltip tip="Image editing"><Icon icon="pen" /></Tooltip>
  </Card>

  <Card title="Grok Imagine">
    `grok-imagine`

    xAI's image model for fast, stylistic generations across many aspect ratios.

    **Prompt** 8K · <Tooltip tip="Aspect ratio"><Icon icon="crop" /></Tooltip> <Tooltip tip="Image editing"><Icon icon="pen" /></Tooltip>
  </Card>

  <Card title="Grok Imagine Quality">
    `grok-imagine-quality`

    The higher-resolution Grok Imagine tier, for crisper detail at 2K.

    **Prompt** 8K · <Tooltip tip="Resolution tier"><Icon icon="up-right-and-down-left-from-center" /></Tooltip> <Tooltip tip="Image editing"><Icon icon="pen" /></Tooltip>
  </Card>

  <Card title="FLUX.2 Pro">
    `flux-2-pro`

    Black Forest Labs' high-fidelity generator with precise pixel-dimension control.

    **Prompt** 10K · <Tooltip tip="Pixel dimensions"><Icon icon="ruler-combined" /></Tooltip> <Tooltip tip="Image editing"><Icon icon="pen" /></Tooltip>
  </Card>

  <Card title="Ideogram V4">
    `ideogram-v4`

    Ideogram's model, strong at typography and graphic design, with a selectable rendering speed.

    **Prompt** 10K · <Tooltip tip="Pixel dimensions"><Icon icon="ruler-combined" /></Tooltip> <Tooltip tip="Image editing"><Icon icon="pen" /></Tooltip>
  </Card>
</CardGroup>

<Tip>
  Not sure where to start? **`nano-banana-pro`** is a strong all-round default,
  **`gpt-image-2`** is great for text in images, and **`seedream-4.5`** is a fast,
  economical photorealistic option.
</Tip>