Nano Banana 2 API
Fast 4K generation with accurate text and search-grounded accuracy.
Nano Banana 2 API playground
Params
Billed at this model’s published rate.
Result
Your generations will appear here
Other models from Google
Same vendor, same API call — the differences are below.
| Model | Use case | Price | Resolution | Duration | Added |
|---|---|---|---|---|---|
| Gemini Omni 1.2 Flash | Text → Video | from $0.0374 / sec | 360p · 720p · 1080p … | 3–15s | Nov 2026 |
| Gemini 4 Argon | Video → Text | from $0.10 / 1M cached in | — | — | Oct 2026 |
| Nano Banana 2.1 | Text → Image | from $1.50 / 1M in | 1K · 2K · 4K | — | Oct 2026 |
| Gemini 3.8 Flash Lite TTS | Text → Speech | from $0.50 / 1M input text tok | — | — | Sep 2026 |
| Gemini 3.8 Flash TTS | Text → Speech | from $0.50 / 1M input text tok | — | — | Sep 2026 |
| Gemini 2.5 Flash | Image → Text | from $0.33 / gen | — | — | Sep 2026 |
| Lyria 3.5 | Music | from $0.088 / gen | — | — | Sep 2026 |
| Gemini 3.8 Flash | Image → Text | from $2.20 / gen | — | — | Sep 2026 |
Nano Banana 2 API in TypeScript, Python and REST
Install the SDK
TypeScript targets Node 20+ and ships with types; Python needs 3.10+ and is async. Over REST there is nothing to install: call https://api.picsart.com from any HTTP client.
npm install @picsart/ai-sdkpip install picsart-ai-sdkAuthenticate
Send your Picsart API key as a bearer token. The SDK does it for you once the client is created.
import { createClient } from '@picsart/ai-sdk';
const ai = createClient({
apiKey: process.env.PICSART_API_KEY, // sent as: Authorization: Bearer <key>
apiUrl: 'https://api.picsart.com',
});import os
from picsart_ai_sdk import create_client
ai = create_client(
api_key=os.environ["PICSART_API_KEY"], # sent as: Authorization: Bearer <key>
api_url="https://api.picsart.com",
)Authorization: Bearer <PICSART_API_KEY>Make a request
Call the model by its id. Over REST, one endpoint serves every model: POST /workflows/v1/models/execute with the model id and its input in the body. The outer params is the envelope every workflow takes; the inner one is this model's input, with the same names generate() takes.
import { createClient } from '@picsart/ai-sdk';
const ai = createClient({
apiKey: process.env.PICSART_API_KEY,
apiUrl: 'https://api.picsart.com',
});
const result = await ai.generate('gemini-3.1-flash-image', {
prompt: "A serene mountain lake at golden hour, ultra detailed",
aspectRatio: "1:1",
resolution: "1K"
});
console.log(result.url); // image URLimport asyncio
import os
from picsart_ai_sdk import create_client
async def main() -> None:
async with create_client(
api_key=os.environ["PICSART_API_KEY"],
api_url="https://api.picsart.com",
) as ai:
result = await ai.generate("gemini-3.1-flash-image", {
"prompt": "A serene mountain lake at golden hour, ultra detailed",
"aspectRatio": "1:1",
"resolution": "1K",
})
print(result.url) # image URL
asyncio.run(main())curl -X POST 'https://api.picsart.com/workflows/v1/models/execute' \
-H 'Authorization: Bearer $PICSART_API_KEY' \
-H 'Content-Type: application/json' \
-d '{
"params": {
"model": "gemini-3.1-flash-image",
"params": {
"prompt": "A serene mountain lake at golden hour, ultra detailed",
"aspectRatio": "1:1",
"resolution": "1K"
}
}
}'Parameters
7 parameters, the same for the SDK and for the inner params over REST. Required ones must be supplied; the rest fall back to their defaults.
| Parameter | Type | Required | Default | Details |
|---|---|---|---|---|
promptWhat to make, in plain language: the subject, style and composition. With an input image or video, what to change. | text | yes | — | — |
aspectRatioShape of the output as width:height. An auto or adaptive option follows the input image. | enum | no | 1:1 | 1:1, 16:9, 9:16, 3:4, 4:3, 3:2, 2:3, 4:5, 5:4, 4:1, 1:4, 8:1, 1:8, 21:9, auto (follows the input) |
resolutionOutput size. Higher costs more and takes longer. | enum | no | 1K | 0.5K, 1K, 2K, 4K |
countHow many results to make in one call. Each one is charged. | enum | no | 1 | 1, 2, 4, 6, 8, 10 |
seedSeed for the random generator. The same seed and inputs repeat a result; leave it out for a random one. | range | no | — | 0–2147483647 (step 1) |
thinkingLevelHow much the model reasons before answering. More is slower and uses more tokens. | enum | no | minimal | minimal (Minimal (faster)), high (High (more reasoning)) |
imageUrlsInput images, as URLs: the image to edit, or references for subject, style or composition. | file (image[]) | no | — | — |
Response
generate() resolves once the job completes. One or more image URLs. Over REST the output is at response.result, inside the gateway's status envelope with the task id, its status and the amount charged.
{
"items": [
{
"url": "https://cdn.picsart.com/…/result"
}
],
"model": "gemini-3.1-flash-image"
}{
"status": "success",
"response": {
"id": "task_abc123",
"status": "COMPLETED",
"result": {
"url": "https://cdn.picsart.com/…/result",
"items": [
{
"url": "https://cdn.picsart.com/…/result"
}
],
"model": "gemini-3.1-flash-image"
}
}
}Nano Banana 2 API advanced usage
Overview
Alongside the unified model APIs, we expose compatibility APIs that take each vendor's original parameters exactly as the vendor defines them — nothing renamed, nothing reshaped on the way through.
That makes them the shortest path onto Picsart if you already work with the vendor directly: the request bodies you've already written keep working as they are. You keep their parameter names and defaults, and you get the vendor's full parameter set rather than the subset that is shared across every model.
The cost is that a request is written for one vendor — switching models later means rewriting it, and results come back in the vendor's own shape. When you'd rather write once and change models freely, use the unified API.
Make a request
Call the workflow with ai.apis.run() in TypeScript or Python, or hit /workflows/{workflow}/execute directly — params is passed through untouched either way.
These endpoints are addressed by workflow name rather than model id: the one in this model's header, since one model runs as one workflow. Authentication is unchanged — your Picsart API key as a bearer token (see Authentication).
import { createClient, ApiRunMode } from '@picsart/ai-sdk';
const ai = createClient({
apiKey: process.env.PICSART_API_KEY,
apiUrl: 'https://api.picsart.com',
});
// Calls the 'gemini/v2/images' workflow directly — params are sent as-is.
const { result, usage } = await ai.apis.run('gemini/v2/images', {
prompt: "A serene mountain lake at golden hour, ultra detailed",
model: "gemini-2.5-flash-image"
}, {
mode: ApiRunMode.SYNC,
});
console.log(result); // workflow-specific output
console.log(usage?.credits); // amount charged, in $import asyncio
import os
from picsart_ai_sdk import ApiRunMode, create_client
async def main() -> None:
async with create_client(
api_key=os.environ["PICSART_API_KEY"],
api_url="https://api.picsart.com",
) as ai:
# Calls the 'gemini/v2/images' workflow directly — params are sent as-is.
response = await ai.apis.run("gemini/v2/images", {
"prompt": "A serene mountain lake at golden hour, ultra detailed",
"model": "gemini-2.5-flash-image",
}, mode=ApiRunMode.SYNC)
print(response["result"]) # workflow-specific output
print((response.get("usage") or {}).get("credits")) # amount charged, in $
asyncio.run(main())curl -X POST 'https://api.picsart.com/workflows/gemini/v2/images/execute' \
-H 'Authorization: Bearer $PICSART_API_KEY' \
-H 'Content-Type: application/json' \
-d '{
"params": {
"prompt": "A serene mountain lake at golden hour, ultra detailed",
"model": "gemini-2.5-flash-image"
}
}'Parameters
9 parameters, sent inside params. These are the vendor's own names, so they line up one-for-one with the vendor's documentation. Required ones must be supplied; the rest fall back to their defaults.
| Parameter | Type | Required | Default | Details |
|---|---|---|---|---|
promptText prompt for image generation | string | yes | — | — |
imageUrlsReference image URLs for editing or style reference | string[] | no | — | — |
aspectRatioOutput aspect ratio. Use "auto" to let the model choose — no aspect ratio is sent to the vendor in that case. | string | no | — | auto1:12:33:23:44:34:55:49:1616:921:91:44:11:88:1 |
imageSizeOutput image resolution/quality. Support is model-dependent: gemini-2.5-flash-image and gemini-3.1-flash-lite-image support 1K only; gemini-3-pro-image supports 1K, 2K, 4K; gemini-3.1-flash-image supports 0.5K, 1K, 2K, 4K. Returns 400 unsupported_image_size for any other value. | string | no | — | 0.5K1K2K4K |
modelGemini image model | string | yes | — | gemini-2.5-flash-imagegemini-3-pro-imagegemini-3-pro-image-previewgemini-3.1-flash-imagegemini-3.1-flash-lite-imagegemini-3.1-flash-image-previewgemini-nano-banana-2.1 |
countNumber of images to generate (1-5) | number | no | 1 | — |
seedSeed for best-effort reproducible output. With count > 1 each image gets its own seed (seed, seed + 1, …) so a batch stays reproducible without returning duplicates. Random when omitted. | number | no | — | 0–2147483647 e.g. 42 |
optionsOptions controlling safety checks and drive integration | object | no | — | — |
└ safety_checksSafety check settings | object | no | — | — |
└ enabledWhether to run content moderation. Defaults to true. | boolean | no | — | — |
└ driveSave result to Picsart Drive | object | no | — | — |
└ nameFile name in Picsart Drive | string | yes | — | — |
└ attributesCustom attributes to attach to the file | object | no | — | — |
└ folderTarget folder in Picsart Drive | object | no | — | — |
└ inputs_transformationInput transformation settings | object | no | — | — |
└ downscale_oversized_imagesWhether to downscale oversized input images. Defaults to false. | boolean | no | — | — |
thinkingConfigConfiguration for the model's thinking/reasoning capabilities. | object | no | — | — |
└ thinkingLevelControls the depth of the model's internal reasoning process. | string | no | — | MINIMALLOWMEDIUMHIGH |
└ thinkingBudgetNumber of thinking tokens the model should use. Range: 0-24576. Set to 0 to disable thinking. | number | no | — | 0–24576 e.g. 1024 |
Response
Over HTTP the output arrives inside a status envelope, at response.result. ai.apis.run() unwraps that envelope for you and resolves to { result, usage } instead. Either way the result itself is the vendor's own shape.
{
"result": {
"imageUrls": [
{
"url": "https://cdn.picsart.com/…/result.jpg",
"mimeType": "image/jpeg",
"driveFile": {
"id": "wf_abc123"
}
}
],
"description": "…"
}
}{
"status": "success",
"response": {
"id": "wf_abc123",
"status": "COMPLETED",
"result": {
"imageUrls": [
{
"url": "https://cdn.picsart.com/…/result.jpg",
"mimeType": "image/jpeg",
"driveFile": {}
}
],
"description": "…"
}
}
}Nano Banana 2 API pricing
Rates
What this model costs per generation, in USD. Every row is an exact rate — the columns are the settings that change it.
| Tier | Use case | Rate |
|---|---|---|
| 0.5K | Any | $0.06 / generation |
| 1K | Any | $0.08 / generation |
| 2K | Any | $0.12 / generation |
| 4K | Image → image | $0.1506 / generation |
| 4K | Text → image | $0.165 / generation |