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

# Generate from a prompt

> Create a fashion image by describing the person, clothing, and scene.

Use `POST /v1/images/generate` when you want to create a new image from a description. No uploaded photo is required.

```bash theme={null}
curl https://api.bitstudio.ai/v1/images/generate \
  -H "Authorization: Bearer $BITSTUDIO_API_KEY" \
  -H 'Content-Type: application/json' \
  -H 'Idempotency-Key: linen-portrait-001' \
  -d '{
    "prompt": "An adult fashion model wearing a cream linen outfit, standing against a warm grey studio background. Full-length portrait, soft daylight, relaxed pose.",
    "aspect_ratio": "3:4"
  }'
```

The omitted settings default to **Nano Banana 2**, standard quality, and **one output image**. This creates clothing from your description. To feature a specific product, [include its reference photos](/api-reference/generate/product-photos).

Save `jobs[0].id`, then [check progress](/api-reference/jobs/get). When the status is `completed`, use `result.url` to display or download the image. Stop polling on `failed` or `policy_blocked` and read `error.message`.

## Structure your directions

Use `model_text` and `outfit_text` for person and clothing descriptions. Put scene, lighting, pose, and composition directions in `prompt`:

```json theme={null}
{
  "model_text": "An adult woman with short dark hair",
  "outfit_text": "A cream linen suit and brown loafers",
  "prompt": "Full-length portrait in a warm grey studio with soft window light, relaxed standing pose",
  "aspect_ratio": "3:4"
}
```

Submit this body to the same generation endpoint. If you need to preserve a specific person from a photo, use [virtual try-on](/api-reference/try-on/single-item).

[Generate with a product →](/api-reference/generate/product-photos) · [Choose output settings →](/api-reference/generate/settings)