Nano-Banana Edit
Google's advanced AI-powered image editing and generation model for intuitive visual transformations.
Overview
Nano-Banana Edit is an image editing model that modifies images using natural language prompts. It combines computer vision and generative AI to perform edits such as object replacement, style transfer, relighting, and background changes while preserving original image context.
Why it looks great
- Natural Language Editing – Modify images using simple text instructions without masking or manual tools.
- Context-Aware Understanding – Accurately interprets scene structure, spatial relationships, and object semantics.
- Style and Tone Preservation – Keeps lighting, shadows, and texture consistent with the original image.
- High Precision Control – Excels at fine-grained edits like color adjustments, object replacement, or composition shifts.
- Creative Versatility – Suitable for concept art, photography, advertising design, and content creation.
Limits and Performance
- Input: existing image (URL) + text prompt.
- Output: edited image in JPEG, PNG, or WEBP format.
- Supports aspect ratios: 1:1, 3:2, 2:3, 3:4, 4:3, 4:5, 5:4, 9:16, 16:9, 21:9.
- Maximum of 10 input images per request.
- NSFW detection is applied to outputs.
Pricing
| Item | Price |
|---|---|
| Per image | $0.038 USD |
Billing Rule
Pricing
is per generated image. Charges apply upon successful generation.
How to Use
- Provide one or more image URLs and a text prompt describing the desired edit.
- Optionally specify aspect ratio and output format.
- The model returns an edited image based on the prompt.
Example prompt: "Replace the cloudy sky with a clear sunset."
Input Parameters
| Parameter | Type | Required | Description |
|---|---|---|---|
| prompt | string | Yes | The positive prompt for the generation. |
| images | array of string | Yes | List of URLs of input images for editing (max 10). |
| aspect_ratio | string | No | Aspect ratio of the generated media. Options: "1:1", "3:2", "2:3", "3:4", "4:3", "4:5", "5:4", "9:16", "16:9", "21:9". |
| output_format | string | No | Format of the output image. Default: "png". Options: "png", "jpeg", "webp". |
Output Format
The response includes:
outputs: Array of URLs to the generated content (empty when status is not completed).status: Task status: created, processing, completed, or failed.urls: Object containing related API endpoints.created_at: ISO timestamp of request creation.has_nsfw_contents: Array of booleans indicating NSFW detection for each output.id: Unique identifier for the prediction.model: Model ID used for the prediction.
Pro tips for best quality
- Use clear, descriptive prompts that specify the desired change.
- Ensure input images have good lighting and resolution.
- For complex edits, break down the instruction into simple steps.
- Avoid prompts that conflict with the original image context.
Note
Please ensure your prompts comply with safety guidelines. If an error occurs, review your prompt for restricted content, adjust it, and try again.