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

# Image Generation

> POST /v1/images/generations — generate images with 25+ models across all styles.

## Endpoint

```
POST https://dolphy.chat/api/v1/images/generations
```

OpenAI-compatible. Use the official `openai` SDK or call directly.

## Request

```json theme={null}
{
  "model": "lustify-v8",
  "prompt": "a dolphin wearing an astronaut helmet, 4k",
  "n": 1,
  "size": "1024x1024",
  "response_format": "url"
}
```

| Field             | Type   | Notes                                                           |
| ----------------- | ------ | --------------------------------------------------------------- |
| `model`           | string | See [`/v1/models`](/models) for the list. Default `lustify-v8`. |
| `prompt`          | string | Up to 5000 chars                                                |
| `n`               | int    | 1–4                                                             |
| `size`            | string | `1024x1024`, `1792x1024`, `1024x1792`, or any `WxH`             |
| `response_format` | enum   | `url` (default, served from our CDN) or `b64_json`              |
| `negative_prompt` | string | Optional — describe what to exclude from the image              |
| `seed`            | int    | Reproducibility                                                 |
| `style_preset`    | string | Provider-specific                                               |

## Response

```json theme={null}
{
  "created": 1730000000,
  "data": [{ "url": "https://firebasestorage.googleapis.com/..." }]
}
```

## Code samples

```ts theme={null}
const r = await client.images.generate({
  model: "lustify-v8",
  prompt: "a cinematic dolphin in space",
  n: 2,
  size: "1024x1024",
});
console.log(r.data.map((d) => d.url));
```

```python theme={null}
r = client.images.generate(
  model="lustify-v8",
  prompt="a cinematic dolphin in space",
  n=2,
  size="1024x1024",
)
print([d.url for d in r.data])
```

## Specialty models

Specialty creative-mode models (`lustify-v8`, `lustify-v7`, `chroma`,
`z-image-turbo`, `wai-Illustrious`) skip provider-side content filtering.
Standard models (Google, OpenAI, etc.) enforce stricter content policies
and may return `422` on rejected prompts. Pick the right model for your
use case.

## Billing

See [Pricing](/pricing). Per-image cost ranges from 1 credit (specialty
models) to 9 credits (Recraft V4 Pro). Multi-image (`n > 1`) bills
linearly.