# Nano Banana 2.1 API (Google)

> Nano Banana 2.1 (gemini-nano-banana-2.1) is Google's newest image model: generation and editing up to 4K with accurate text, up to 14 input images and 14 aspect ratios, from about 2 cents an image.

Page: https://cheapergen.com/models/google/nano-banana-2.1 · API reference: https://cheapergen.com/models/google/nano-banana-2.1/api

## Pricing
Official list price source: Gemini API list price for gemini-nano-banana-2.1 ($30 / 1M image output tokens: $0.0336 at 1K, $0.0504 at 2K, $0.1134 at 4K; input $1.50 / 1M, thinking $7.50 / 1M tokens). Prices verified 2026-10-16.

| Tier | Official | Cheaper Gen. | Save |
|---|---|---|---|
| 1K | $0.0416/image | $0.0220/image | 47% |
| 2K | $0.0584/image | $0.0330/image | 43% |
| 4K | $0.1214/image | $0.0495/image | 59% |

## Specs
- Resolutions: 1K · 2K · 4K
- Aspect ratios: 1:1 · 2:3 · 3:2 · 3:4 · 4:3 · 4:5 · 5:4 · 9:16 · 16:9 · 21:9 · 1:4 · 4:1 · 1:8 · 8:1
- Edit inputs: Up to 14 images
- Model id (Google): gemini-nano-banana-2.1

## Endpoints

### Text to Image — `POST https://cheapergen.com/api/v1/run/google/nano-banana-2.1/text-to-image`

| Param | Type | Required | Description |
|---|---|---|---|
| prompt | string | yes | Describe the shot: subject, action, camera movement, lighting and style. |
| aspect_ratio | string (enum) (1:1, 2:3, 3:2, 3:4, 4:3, 4:5, 5:4, 9:16, 16:9, 21:9, 1:4, 4:1, 1:8, 8:1) | no | Output image shape. |
| resolution | string (enum) (1K, 2K, 4K) | no | Output size: 1K ≈ 1 megapixel, 2K ≈ 4 MP, 4K ≈ 16 MP. |

```bash
curl -X POST https://cheapergen.com/api/v1/run/google/nano-banana-2.1/text-to-image \
  -H "Authorization: Bearer $CHEAPERGEN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "prompt": "A cozy ramen shop at night in the rain, warm light spilling onto wet pavement, a neon sign that reads “OPEN”, photorealistic",
  "aspect_ratio": "1:1",
  "resolution": "1K"
}'
```

### Image Edit — `POST https://cheapergen.com/api/v1/run/google/nano-banana-2.1/edit`

| Param | Type | Required | Description |
|---|---|---|---|
| prompt | string | yes | Describe the shot: subject, action, camera movement, lighting and style. |
| image_urls | string[] (URLs) | yes | Up to 14 images to edit or combine — products, people, styles. Refer to them as "image 1", "image 2"… in the prompt. |
| aspect_ratio | string (enum) (1:1, 2:3, 3:2, 3:4, 4:3, 4:5, 5:4, 9:16, 16:9, 21:9, 1:4, 4:1, 1:8, 8:1) | no | Output image shape. |
| resolution | string (enum) (1K, 2K, 4K) | no | Output size: 1K ≈ 1 megapixel, 2K ≈ 4 MP, 4K ≈ 16 MP. |

```bash
curl -X POST https://cheapergen.com/api/v1/run/google/nano-banana-2.1/edit \
  -H "Authorization: Bearer $CHEAPERGEN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "prompt": "Put the product from image 1 on the marble counter in image 2, matching the soft morning light",
  "image_urls": [
    "https://cdn.cheapergen.ai/samples/portrait.jpg"
  ],
  "aspect_ratio": "1:1",
  "resolution": "1K"
}'
```

## Getting the result
Jobs are async. Poll `GET https://cheapergen.com/api/v1/generations/{id}` until status is succeeded/failed/canceled, or pass `webhook_url`. Failed generations are refunded automatically.

## FAQ
**How is Nano Banana 2.1 priced?**
Nano Banana 2.1 is billed per generated image at Google's rates for the output resolution; the prompt and each input image add a small amount of input tokens. Failed or blocked generations are not charged.

**Can it edit my own images?**
Yes — use Image Edit with one or more input images and describe the change in plain language. It can combine several images (products, people, places) into one scene and keeps details consistent.

**Is this the official Nano Banana 2.1?**
Yes. Requests run on Google's Gemini image model through an official Google channel — the same model as the Gemini API, at a lower price.

**Can I use the output commercially?**
Yes, subject to Google's Generative AI terms and our Terms of Service. Images carry Google's invisible SynthID watermark.

Nano Banana 2.1 is developed by Google. Cheaper Gen. provides API access and does not own the model.