# Scene · Nano Banana Pro (scene)

> A scene is the identity + product composite still that later becomes a video first frame. This endpoint makes one scene PNG with Nano Banana Pro and returns a `scene_id` that /v1/ugc/* and /v1/video/* accept.

Status: roasOS API v1.1. Catalog `model-catalog@2026-10-06.v5`, pricing `customer-pricing@2026-11-05.v15`. The endpoint `POST /v1/scene/nano-banana-pro` is live. `POST /v1/generations` has no `scene` type, so this endpoint is the only path to this model.

## Endpoint
POST https://api.roasos.com/v1/scene/nano-banana-pro
Auth: Authorization: Bearer roasos_sk_… (scope api:write). Idempotency-Key required. quote_only / quote_id / webhook_url as in https://docs.roasos.com/llms.txt.

## Good at
Use this endpoint to put an avatar (identity) and a product into one framed still, ready to animate. The worker adds the CHARACTER PLACEMENT block (with the rule "re-render the whole person, never paste"), the PRODUCT PLACEMENT block, and the STYLE GUIDANCE block for any references, so the model knows which image is the person, which is the product, and which is mood only. The default frame is `9:16` for vertical ad placements. Set `remix: true` with references to restyle an existing ad frame. The output is a scene record, not a plain image: it carries a `scene_id`. It does not make 4K output; the scene is a first frame, and video models read at most 2K.

## Pricing
Unit: scene. Pricing key: `nano-banana-pro`.

| request | credits |
|---|---|
| `1K` | 2 |
| `2K` | 2 |

Credits come from the roasOS pricing table (`MODEL_COSTS`) through the catalog. No other field changes the price.

## Modes
- `composite`
Provider endpoints: `fal-ai/nano-banana-pro` · `fal-ai/nano-banana-pro/edit`.

## Request fields
| field | type | required | default | rule |
|---|---|---|---|---|
| `aspect_ratio` | enum | no | `9:16` | `auto`, `21:9`, `16:9`, `3:2`, `4:3`, `5:4`, `1:1`, `4:5`, `3:4`, `2:3`, `9:16` |
| `avatar_id` | string | no | — | UUID |
| `display_prompt` | string | no | — | ≥ 1 character; ≤ 500 characters |
| `name` | string | no | — | ≥ 1 character; ≤ 200 characters |
| `product_id` | string | no | — | UUID |
| `product_images` | string[] | no | — | ≤ 1 items |
| `prompt` | string | yes | — | ≥ 1 character; ≤ 8,000 characters |
| `quote_id` | string | no | — | UUID |
| `quote_only` | boolean | no | `false` | — |
| `references` | string[] | no | — | ≤ 12 items |
| `remix` | boolean | no | `false` | — |
| `resolution` | enum | no | `1K` | `1K`, `2K` |
| `wait_timeout_seconds` | integer | no | `0` | 0–60 |
| `webhook_url` | string | no | — | https URL; ≤ 4,096 characters |

The schema is strict: a field that is not in this table returns `400 unsupported_option`. Field names are the roasOS role names, one name per role across every family.

## Media
- `avatar_id` — identity; the saved character reference sheet, with a derived portrait only where the model requires a single face or source frame
- `product_id` — product; resolves to the product image list
- `product_images` — product reference images
- `references` — style and appearance references

| role | items | max size | content types | duration |
|---|---|---|---|---|
| `product_images` | ≤ 1 | 10 MB | `image/jpeg`, `image/png`, `image/webp` | — |
| `references` | ≤ 12 | 10 MB | `image/jpeg`, `image/png`, `image/webp` | — |

- Refused: `elements`, `reference_images`, `image_url`, `end_image_url`, `scene_id`, `scene_image`, `video_urls`, `video_url`, `audio_urls`, `audio_url`, `voice_id`. A refused role returns `400 unsupported_option`.
- Every URL must be an immutable object on `https://cdn.roasos.com/…` that your workspace owns; any other URL is `400 reference_media_not_workspace_owned`. Upload with `POST /v1/uploads` (direct POST of the bytes to `put_url`, or a `source_url` import), then use the returned CDN URL.

## Prompt
Limit: 8,000 characters for the caller; the wrapper counts toward the 10,000-character engine cap. No tag syntax; the wrapper names the roles.
- Describe the frame you want to animate: where the person stands, where the product is, what the person does, the setting and the light.
- Give the framing that the video model needs ("half body, product at chest height, eyes to camera").
- Do not describe the person's face or the product's label; the roles carry them and the wrapper tells the model to re-render, never paste.
- Leave room in the frame for motion (hands, product lift).
- Use `references` for mood and palette only; state "match the lighting of the references" in the prompt if that is the intent.
- Keep the same prompt structure across scenes of one campaign so the frames match.

## Not exposed
- `output_format`: pinned to `png`.
- `safety_tolerance`: pinned to `4`.
- `enable_web_search`: pinned to `false`.
- `limit_generations`: pinned to `true`.
- `system_prompt`: not accepted by this public endpoint.
- `sync_mode`: not accepted by this public endpoint.
- `num_images`: pinned to `1`.
- `resolution`: values not exposed — `4K`.

## Result
Result kind `scene`. Always present: `scene_id`, `url`.
```json
{ "data": { "id": "gen_…", "kind": "scene", "model": "nano-banana-pro", "status": "completed", "result": {"scene_id":"00000000-0000-4000-8000-000000000000","url":"https://cdn.roasos.com/workspaces/ws_01/result"} } }
```
Files live on `https://cdn.roasos.com` and are immutable; a result URL can be a media input to the next call. Poll `GET /v1/jobs/{id}/status` → `{ id, status, updated_at }`, then `GET /v1/jobs/{id}/result` (`409 run_not_completed` until the job is `completed`, `failed` or `canceled`). Statuses: `queued → processing → completed | failed | canceled`.

## Example
```bash
# 1. quote — nothing is spent
curl -X POST https://api.roasos.com/v1/scene/nano-banana-pro \
  -H "Authorization: Bearer roasos_sk_live_…" \
  -H "Idempotency-Key: 0f1c9e2a-nano-banana-pro-0001" \
  -H "Content-Type: application/json" \
  -d '{"quote_only":true,"prompt":"A woman holds the bottle at eye level in a bright kitchen, morning light, static camera."}'
# → 200 { "data": { "kind": "quote", "quote_id": "q_7Hk…", "credits": 2, "expires_at": "…", "confirmation": { "model": "nano-banana-pro", "example": "1K" } } }

# 2. run — the same body and the same Idempotency-Key, quote_id instead of quote_only
curl -X POST https://api.roasos.com/v1/scene/nano-banana-pro \
  -H "Authorization: Bearer roasos_sk_live_…" \
  -H "Idempotency-Key: 0f1c9e2a-nano-banana-pro-0001" \
  -H "Content-Type: application/json" \
  -d '{"quote_id":"q_7Hk…","prompt":"A woman holds the bottle at eye level in a bright kitchen, morning light, static camera."}'
# → 201 { "data": { "id": "gen_…", "kind": "scene", "model": "nano-banana-pro", "status": "queued", "credits_reserved": 2, "links": { "status": "/v1/jobs/gen_…/status", "result": "/v1/jobs/gen_…/result" } } }

# 3. poll
curl https://api.roasos.com/v1/jobs/gen_…/status -H "Authorization: Bearer roasos_sk_live_…"
curl https://api.roasos.com/v1/jobs/gen_…/result -H "Authorization: Bearer roasos_sk_live_…"
```

## Today
POST /v1/scenes `{ "name" (≤ 200), "prompt" (≤ 10,000), "model": "nano-banana-pro", "resolution": "1K"|"2K"|"4K", "webhook_url"?, "quote_only"?, "quote_id"?, "options": { "aspect_ratio", "reference_images" (≤ 4), "avatar_id", "product_image", "mode": "ugc"|"product"|"custom" } }`. Caution: today's public route sends the prompt raw, with no CHARACTER / PRODUCT / STYLE placement wrapper; today's default `aspect_ratio` is `16:9`; `3:2` and `2:3` are advertised but refused. POST /v1/scenes becomes an alias of this endpoint.

## See also
https://docs.roasos.com/models/scene/nano-banana-pro · https://api.roasos.com/v1/catalog/scene/nano-banana-pro · fal-ai/nano-banana-pro · fal-ai/nano-banana-pro/edit