Nano Banana · GPT Image · MiniMax video

Image and video generation, from $0.006 a picture.

One API key for Nano Banana 2, Nano Banana Pro, GPT Image 2.5 and MiniMax H3 video. Top up by card, pay per result, and never pay for a failed or blocked generation.

No subscription. Free starter credit when you confirm your email.

POST /v1/api/generate
curl https://api.xelentapi.com/v1/api/generate \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "nano-banana-2",
    "prompt": "a ceramic mug on a marble counter",
    "aspectRatio": "4:3",
    "imageSize": "2K"
  }'
succeeded {"id":"gen_7Qm…","status":"succeeded","results":[{"url":"https://api.xelentapi.com/files/…/0.png"}]}

image

Nano Banana 2 Lite

Gemini 3.1 Flash Lite Image. Fastest and cheapest; good for drafts and volume.

$0.006 per image

nano-banana-2-lite

image

Nano Banana 2

Gemini 3.1 Flash Image. Balanced quality and consistency, 1K to 4K at one price.

$0.016 per image

nano-banana-2

image

Nano Banana Pro

Gemini 3 Pro Image. Highest quality and complex edits, 1K to 4K at one price.

$0.025 per image

nano-banana-pro

image

GPT Image 2

GPT Image 2 at 1K. Strong text rendering and prompt following.

$0.008 per image

gpt-image-2

image

GPT Image 2.5 Sunburst

GPT Image 2.5 up to 4K with quality up to max, transparent backgrounds.

$0.034 per image

gpt-image-2.5-sunburst

video

MiniMax H3 Video

Text, image and audio to video. 1 to 15 seconds; 1080p up to 10 seconds. Billed per second.

$0.014/s at 480p · $0.020/s at 768p · $0.042/s at 1080p

minimax-h3

You only pay for results

Credit is held when a job starts and charged only when it succeeds. Failures, timeouts and content-policy blocks are released back to your balance automatically.

Drop-in compatible

One simple JSON endpoint for every model, plus an OpenAI-compatible /v1/images/generations. Existing OpenAI image code only needs a new base URL and key.

Blocking, streaming or async

Wait for the image in one call, stream progress as server-sent events, or submit and poll. Your choice per request.

Keys with spending limits

Create a key per app or per customer, cap what each can spend, and revoke one without touching the others.

Every request logged

See each generation with its model, cost, status and output in your dashboard, and your spend by day and by model.

Your files, on a fast CDN

Outputs are served from our edge storage for 7 days, long enough to download them into your own storage.

Questions

How does billing work?

You add credit by card from $5. Each successful image or second of video is deducted at the listed price. There is no subscription and no monthly minimum.

What happens if a generation fails?

You are not charged. The credit reserved for it returns to your balance as soon as the failure is known.

Can I use it from my existing code?

Yes. If you use the OpenAI SDK, point base_url at https://api.xelentapi.com/v1 and call images.generate.

What content is allowed?

Everything our acceptable use policy allows. Requests are moderated, and accounts that repeatedly try to generate prohibited content are closed.

Start generating in two minutes.

Create a free account