Nano Banana Pro Image API

Overview

Model: Nano Banana Pro; supports text-to-image and up to 8 reference images (30MB each).
Resolution: 1K / 2K / 4K; default 1K (higher resolutions cost more).
Aspect ratios: 1:1, 16:9, 9:16, 21:9, plus others and auto.
Output: png / jpg / jpeg (default png).
Async API: submit POST https://api.wizzx.ai/api/v1/task/submit/nano-banana-pro, poll POST https://api.wizzx.ai/api/v1/task/status.

Breaking Change (2025-12-10): image_input replaced by image_urls array.

Auth

Authorization: Bearer <your_api_key>

Submit job

Endpoint: POST https://api.wizzx.ai/api/v1/task/submit/nano-banana-pro, Content-Type: application/json

Field	Type	Required	Default/Limit	Description
`prompt`	string	Yes	≤10000 UTF-8 bytes	Text prompt
`image_urls`	array<string>	No	Max 8, ≤30MB each	Reference images (JPEG/PNG/WebP)
`aspect_ratio`	string	No	Default `1:1`	`1:1`, `2:3`, `3:2`, `3:4`, `4:3`, `4:5`, `5:4`, `9:16`, `16:9`, `21:9`, `auto`
`resolution`	string	No	Default `1K`	`1K` / `2K` / `4K`
`output_format`	string	No	Default `png`	`png` / `jpg` / `jpeg`

Use clear, consistent reference images (1-3 typical, up to 8).
Aspect ratio hints composition; auto lets the model pick.

Example (text-to-image)

curl -X POST https://api.wizzx.ai/api/v1/task/submit/nano-banana-pro \
  -H "Authorization: Bearer <api_key>" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "A futuristic city at night with neon lights",
    "aspect_ratio": "16:9",
    "resolution": "2K",
    "output_format": "png"
  }'

Example (multi-reference / style transfer)

curl -X POST https://api.wizzx.ai/api/v1/task/submit/nano-banana-pro \
  -H "Authorization: Bearer <api_key>" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "Transform this into anime style",
    "image_urls": [
      "https://example.com/source.jpg",
      "https://example.com/style2.jpg"
    ],
    "aspect_ratio": "1:1",
    "resolution": "4K",
    "output_format": "png"
  }'

Query status

Endpoint: POST https://api.wizzx.ai/api/v1/task/status

{
  "task_id": "4bbbc9e0-8de8-4f35-a31e-8596c1a6d18a"
}

Client-visible states are PENDING, QUEUED, PROCESSING, REVIEW_REQUIRED, SUCCEEDED, and FAILED. See the task status reference.

Pricing

Resolution	Credits per image
1K / 2K	108
4K	144

Read /pricing immediately before submission for current tiers.

Rate limits

Rate limits can change. On HTTP 429, honor retry_after or X-Retry-After and do not automatically resubmit an ambiguous paid request.

Migration (image_input → image_urls)

Old (deprecated):

{
  "prompt": "...",
  "image_input": [
    "https://example.com/img1.jpg"
  ]
}

New:

{
  "prompt": "...",
  "image_urls": [
    "https://example.com/img1.jpg"
  ]
}

Best practices

Write prompts with subject, scene, lighting, style (e.g., photorealistic, digital art).
Iterate with 1K/2K, finalize with 4K.
Choose aspect ratio for target surface: 9:16 for portrait, 16:9/21:9 for landscape, 1:1 for covers; use auto if unsure.
Keep references 1-3, consistent in style and resolution; avoid low-res or overexposed images.

Getting started

Agent integrations

Video Generation

Image Generation

Nano Banana Pro Image API

Overview

Auth

Submit job

Example (text-to-image)

Example (multi-reference / style transfer)

Query status

Pricing

Rate limits

Migration (image_input → image_urls)

Best practices

​Overview

​Auth

​Submit job

​Example (text-to-image)

​Example (multi-reference / style transfer)

​Query status

​Pricing

​Rate limits

​Migration (image_input → image_urls)

​Best practices

Overview

Auth

Submit job

Example (text-to-image)

Example (multi-reference / style transfer)

Query status

Pricing

Rate limits

Migration (image_input → image_urls)

Best practices