Text-to-image
Elo · AA-Image-T2I v2.1
Create images and refine them with a prompt. Choose up to 4K output, bring reference images, and keep text and visual details consistent.
Nano Banana 2.1 brings image generation and conversational editing into the same model. The update emphasizes text rendering and consistency across edits, making it useful to evaluate for layouts and recurring visual concepts as well as individual images.
Start from a written brief, or supply up to five reference images when the request needs existing visual material. Teams use it to explore alternative scenes for a campaign before final production, and to test headlines and labels in context. Inspect spelling and placement in every generated result before it ships.
The API supports square, landscape and portrait aspect ratios, with 1K, 2K and 4K output. Use the framing required by your design rather than cropping every result later. Optional web search can add current information when the brief depends on real-world subjects.
Compare Nano Banana 2.1 on image quality, editing and creator API price.
Elo · AA-Image-T2I v2.1
Elo · AA-Image-Editing v2.1
USD per 1,000 images
List price on each model creator’s API at 1024×1024, not the AI/ML API rate. Our rates are in Pricing
Compare five models across image generation and editing.
Blind pairwise human votes on images generated from text prompts.
Blind pairwise human votes on instruction-based edits of an input image.
| Usage | Price | Unit |
|---|---|---|
| 1K generation | $0.104 | image · standard |
| 2K generation | $0.156 | image · 1.5× |
| 4K generation | $0.208 | image · 2× |
Creator API rates per 1,000 images at 1024×1024; AI/ML API pricing is listed above.
| Model | Creator API price |
|---|---|
| $33.60 | |
| $38.90 | |
| $210.70 | |
| $210.70 | |
| $211.00 |
Give your agent the model details, code and documentation.
export AIMLAPI_KEY="8f2b4c7d9a1e6b3f…"curl 'https://api.aimlapi.com/v1/images/generations' \
-H "Authorization: Bearer $AIMLAPI_KEY" \
-H 'Content-Type: application/json' \
-d '{
"model": "google/gemini-nano-banana-2.1",
"prompt": "A minimal ceramic vase on a sunlit stone table, editorial product photography.",
"resolution": "2K",
"aspect_ratio": "1:1"
}'import os
import requests
headers = {"Authorization": "Bearer " + os.environ["AIMLAPI_KEY"]}
payload = {
"model": "google/gemini-nano-banana-2.1",
"prompt": "A minimal ceramic vase on a sunlit stone table, editorial product photography.",
"resolution": "2K",
"aspect_ratio": "1:1"
}
response = requests.post(
"https://api.aimlapi.com/v1/images/generations",
headers=headers,
json=payload,
timeout=120,
)
response.raise_for_status()
data = response.json()
print(data["data"][0]["url"])async function main() {
const headers = {
Authorization: "Bearer " + process.env.AIMLAPI_KEY,
"Content-Type": "application/json",
};
const response = await fetch(
"https://api.aimlapi.com/v1/images/generations",
{
method: "POST",
headers,
body: JSON.stringify({
"model": "google/gemini-nano-banana-2.1",
"prompt": "A minimal ceramic vase on a sunlit stone table, editorial product photography.",
"resolution": "2K",
"aspect_ratio": "1:1"
}),
},
);
if (!response.ok) throw new Error(await response.text());
const data = await response.json();
console.log(data.data[0].url);
}
main().catch(error => {
console.error(error);
process.exitCode = 1;
});The API reference lists 1K, 2K and 4K resolutions, with an adjustable aspect ratio.
Yes. The AI/ML API request schema accepts up to five image URLs alongside the prompt.
Optional search grounding can help with subjects that need current information.
The published standard rate is $0.104 per generation. 2K uses a 1.5× multiplier; 4K uses 2×.