AI Image Generation API: Every Top Model, One Endpoint
TL;DR: GenRelay gives you a single, OpenAI-compatible REST API to call Google's Nano Banana and Gemini image models plus OpenAI's GPT Image 2 — no separate API keys, no separate billing, no separate SDKs per provider. Pay-as-you-go in USD, starting at $0.02/image.
Supported image models
| Model | Provider | Price |
|---|---|---|
| Nano Banana 2 | $0.02 / image | |
| Gemini 3.1 Flash Image | $0.02 / image | |
| OpenAI GPT Image 2 | OpenAI | $0.02 / image |
| Nano Banana Pro | $0.03 / image | |
| Gemini 3 Pro Image | $0.03 / image |
Rates as of 2026-09-24 — see the live pricing page for current numbers.
Quick start
POST https://genrelay.ai/v1/images/generations
Authorization: Bearer YOUR_KEY
Content-Type: application/json
{
"model": "nano-banana-2",
"prompt": "a studio product photo of a ceramic mug on a white background"
}
The API is OpenAI-compatible — if you already have code calling OpenAI's image endpoint, switching to GenRelay is usually a one-line change: swap the base URL and API key.
Why one endpoint instead of five
- One API key, one bill, one rate-limit policy — instead of separate accounts with Google, OpenAI, and every other provider.
- Swap models by changing a single string parameter, so you can A/B test quality/cost across providers without rewriting integration code.
- Full commercial license on generated assets, with a $0.50 free credit to test before you commit spend.
Common use cases
E-commerce product photography, headshot generation, jewelry and eyewear product shots, and other programmatic image generation workflows — see the blog for industry-specific integration guides.