# Nano Banana 2 Image Editing

> Edit images with text prompts. Make targeted changes like adding or removing objects, changing styles, or modifying specific elements while preserving the rest of the image.

## Overview

- **Endpoint**: `https://queue.modelrunner.run/google/nano-banana-2/edit`
- **Model ID**: `google/nano-banana-2/edit`
- **Category**: image-to-image
- **Kind**: inference
- **Tags**: none

## Pricing

- **up to 0.5 megapixels**: $0.045
- **up to 1.5 megapixels**: $0.067
- **up to 5 megapixels**: $0.101
- **up to 18 megapixels**: $0.151

## Request Lifecycle

This model runs on the ModelRunner **asynchronous queue API** — a single POST does not return the output. Every call requires an `Authorization: Key $MODEL_RUNNER_KEY` header. Run three steps:

1. **Submit** — `POST https://queue.modelrunner.run/google/nano-banana-2/edit` with a JSON body holding the input fields at the top level. The body may also include a reserved top-level `metadata` object — a flat string map (max 16 keys, key ≤64 / value ≤512 chars) stored on the request for your own tagging. It is never sent to the model; filter your request history with `GET https://queue.modelrunner.run/requests?metadata=<url-encoded JSON>` (exact key=value matches, AND-ed). The response carries request handles only (no output yet):

```json
{
  "status": "IN_QUEUE",
  "request_id": "<21-char id>",
  "status_url": "https://queue.modelrunner.run/google/nano-banana-2/edit/requests/<request_id>/status",
  "response_url": "https://queue.modelrunner.run/google/nano-banana-2/edit/requests/<request_id>",
  "cancel_url": "https://queue.modelrunner.run/google/nano-banana-2/edit/requests/<request_id>/cancel"
}
```

2. **Poll status** — `GET <status_url>` until `status` is `COMPLETED`. Possible values are `IN_QUEUE`, `IN_PROGRESS`, `COMPLETED`, `FAILED`, `CANCELLED`. A `FAILED` request responds with HTTP 400 and an `error` field.

3. **Read result** — `GET <response_url>`. Returns the finished request, including the generated `output`:

```json
{ "id": "<request_id>", "status": "COMPLETED", "output": ..., "input": ... }
```

The JavaScript and Python SDKs below perform steps 2–3 for you. In any language without an SDK (Swift, Go, Kotlin, etc.) you must implement the polling loop and the final result fetch yourself — see the cURL example for the full flow.

### Input Schema

- **`images`** (`array<string>`, _required_):
  The input image(s) for editing. The model supports providing multiple images in a single request.

- **`prompt`** (`string`, _required_):
  The text prompt for image editing.

- **`image_size`** (`ImageSize | image_size_enum`, _optional_):
  The size of the generated image. Use a preset string (e.g. '4_3_1k') or a custom {width, height} object.
  - Default: `"1_1_2k"`
  - Options: `"1_1_512"`, `"1_1_1k"`, `"1_1_2k"`, `"1_1_4k"`, `"16_9_512"`, `"16_9_1k"`, `"16_9_2k"`, `"16_9_4k"`, `"9_16_512"`, `"9_16_1k"`, `"9_16_2k"`, `"9_16_4k"`, `"4_3_512"`, `"4_3_1k"`, `"4_3_2k"`, `"4_3_4k"`, `"3_4_512"`, `"3_4_1k"`, `"3_4_2k"`, `"3_4_4k"`, `"3_2_512"`, `"3_2_1k"`, `"3_2_2k"`, `"3_2_4k"`, `"2_3_512"`, `"2_3_1k"`, `"2_3_2k"`, `"2_3_4k"`, `"4_5_512"`, `"4_5_1k"`, `"4_5_2k"`, `"4_5_4k"`, `"5_4_512"`, `"5_4_1k"`, `"5_4_2k"`, `"5_4_4k"`, `"4_1_512"`, `"4_1_1k"`, `"4_1_2k"`, `"4_1_4k"`, `"1_4_512"`, `"1_4_1k"`, `"1_4_2k"`, `"1_4_4k"`, `"8_1_512"`, `"8_1_1k"`, `"8_1_2k"`, `"8_1_4k"`, `"1_8_512"`, `"1_8_1k"`, `"1_8_2k"`, `"1_8_4k"`, `"21_9_512"`, `"21_9_1k"`, `"21_9_2k"`, `"21_9_4k"`, `"auto"`

- **`safety_settings`** (`array<object>`, _optional_):
  A list of unique safety settings for blocking unsafe content. Supported categories: HARM_CATEGORY_HARASSMENT, HARM_CATEGORY_HATE_SPEECH, HARM_CATEGORY_SEXUALLY_EXPLICIT, HARM_CATEGORY_DANGEROUS_CONTENT. Supported thresholds: HARM_BLOCK_THRESHOLD_UNSPECIFIED, BLOCK_LOW_AND_ABOVE, BLOCK_MEDIUM_AND_ABOVE, BLOCK_ONLY_HIGH, BLOCK_NONE. This setting can only be configured via the API.
  - Default: `\[{"category":"HARM_CATEGORY_HARASSMENT","threshold":"BLOCK_LOW_AND_ABOVE"},{"category":"HARM_CATEGORY_HATE_SPEECH","threshold":"BLOCK_LOW_AND_ABOVE"},{"category":"HARM_CATEGORY_SEXUALLY_EXPLICIT","threshold":"BLOCK_LOW_AND_ABOVE"},{"category":"HARM_CATEGORY_DANGEROUS_CONTENT","threshold":"BLOCK_LOW_AND_ABOVE"}\]`

- **`google_search_grounding`** (`boolean`, _optional_):
  Enables Google Search grounding to provide more accurate and up-to-date information in the generated image.
  - Default: `false`

### Output Schema

_No `Output` schema properties are available._

## Default Example

**Input**

```json
{
  "images": [
    "https://media.modelrunner.ai/yFVc9Q0GFwVBAZthX2ra1.png"
  ],
  "prompt": "Change the empty wooden park bench to an ornate, dark green wrought-iron bench. Keep the lighting, perspective, and autumn leaves exactly the same.",
  "image_size": "16_9_4k",
  "safety_settings": [],
  "google_search_grounding": false
}
```

**Output**

```json
"https://media.modelrunner.ai/TVR6j2kFQxuos7rsepY6M.png"
```
## Usage Examples

### cURL

The queue API is asynchronous: submit the request, poll `status_url` until it is `COMPLETED`, then read the result from `response_url`. Requires `jq`.

```bash
# 1. Submit the request (returns request handles, not the output)
SUBMIT=$(curl --silent --request POST \
  --url https://queue.modelrunner.run/google/nano-banana-2/edit \
  --header "Authorization: Key $MODEL_RUNNER_KEY" \
  --header "Content-Type: application/json" \
  --data '{
  "images": [
    "https://media.modelrunner.ai/yFVc9Q0GFwVBAZthX2ra1.png"
  ],
  "prompt": "Change the empty wooden park bench to an ornate, dark green wrought-iron bench. Keep the lighting, perspective, and autumn leaves exactly the same.",
  "image_size": "16_9_4k",
  "safety_settings": [],
  "google_search_grounding": false
}')

STATUS_URL=$(echo "$SUBMIT" | jq -r '.status_url')
RESPONSE_URL=$(echo "$SUBMIT" | jq -r '.response_url')

# 2. Poll until the request leaves the queue / in-progress state
while true; do
  STATUS=$(curl --silent --url "$STATUS_URL" \
    --header "Authorization: Key $MODEL_RUNNER_KEY" | jq -r '.status')
  echo "Status: $STATUS"
  case "$STATUS" in
    COMPLETED) break ;;
    FAILED|CANCELLED) echo "Request $STATUS"; exit 1 ;;
  esac
  sleep 1
done

# 3. Read the finished request, including the generated output
curl --silent --url "$RESPONSE_URL" \
  --header "Authorization: Key $MODEL_RUNNER_KEY"
```

### JavaScript

```javascript
import { modelrunner } from "@modelrunner/client";

const result = await modelrunner.subscribe("google/nano-banana-2/edit", {
  input: {
  "images": [
    "https://media.modelrunner.ai/yFVc9Q0GFwVBAZthX2ra1.png"
  ],
  "prompt": "Change the empty wooden park bench to an ornate, dark green wrought-iron bench. Keep the lighting, perspective, and autumn leaves exactly the same.",
  "image_size": "16_9_4k",
  "safety_settings": [],
  "google_search_grounding": false
}
});

console.log(result.data);
```

### Python

```python
import asyncio
import modelrunner_ai

async def main():
    response = await modelrunner_ai.submit_async(
        "google/nano-banana-2/edit",
        arguments={
  "images": [
    "https://media.modelrunner.ai/yFVc9Q0GFwVBAZthX2ra1.png"
  ],
  "prompt": "Change the empty wooden park bench to an ornate, dark green wrought-iron bench. Keep the lighting, perspective, and autumn leaves exactly the same.",
  "image_size": "16_9_4k",
  "safety_settings": [],
  "google_search_grounding": false
}
    )
    result = await response.get()
    print(result["output"])

asyncio.run(main())
```

## Additional Resources

- [Playground](https://modelrunner.ai/models/google/nano-banana-2/edit)
- [OpenAPI Schema](https://modelrunner.ai/models/google/nano-banana-2/edit/openapi.json)
- [LLM Instructions](https://modelrunner.ai/models/google/nano-banana-2/edit/llms.txt)