Nano Banana API
The latest Nano Banana API, powered by Google Gemini 3.1 Flash Image Preview for fast 4K generation and editing with up to 14 reference images.
Access Nano Banana API through APIMaster's OpenAI-compatible images endpoint. Nano Banana 2 supports text-to-image and image-to-image with extreme aspect ratios, optional Google search grounding, and resolution tiers from 0.5K to 4K, billed per image.
What is the Nano Banana API?
Nano Banana API on APIMaster is powered by Nano Banana 2, Google's gemini-3.1-flash-image model. It delivers fast multimodal image generation with strong prompt following, up to 14 reference images for editing workflows, and native 4K output. Use the sync endpoint for one-shot results or async submit plus poll for batch pipelines.
Channel 93
Fingerprint history
No detection data
Uptime
No detection data
Channel 92
Fingerprint history
No detection data
Uptime
No detection data
Channel 72
Fingerprint history
No detection data
Uptime
No detection data
Transparent per-image pricing by resolution — pay only for the images you generate.
Nano Banana API — Pricing
| Model | Our price | Official | Save |
|---|---|---|---|
| Nano Banana 2 | $0.045/img | $0.0672/img | 33% |
Capabilities
Native 4K output
Generate up to 4K resolution images with sharp detail — resolution billed in 0.5K / 1K / 2K / 4K tiers.
Text & image-to-image
Up to 14 reference images for editing, style transfer, and multi-image fusion workflows.
Extreme aspect ratios
Supports auto sizing plus ratios from 1:1 through 21:9 and extreme 1:4 / 4:1 — ideal for banners and social formats.
Use cases
Quick start
curl https://apimaster.ai/v1/images/generations \
-H "Authorization: Bearer $APIMASTER_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "gemini-3.1-flash-image",
"prompt": "An orange cat at sunset, watercolor"
}'Nano Banana API — Frequently Asked Questions
Which model id do I use for Nano Banana API?
Always use gemini-3.1-flash-image in the model field. Nano Banana 2 is the latest Nano Banana API product on APIMaster.
Which endpoint do I call?
POST /v1/images/generations (sync) or /v1/images/generations/async — not /chat/completions. See Quick start below.
How is billing calculated?
Per image, multiplied by resolution tier (0.5K/1K base, 2K ≈ 4/3×, 4K ≈ 2×). See the price table and usage logs for exact charges.
Can I use reference images?
Yes — up to 14 reference images via image_urls or multipart upload in the playground. JPEG, PNG, WebP, max 10MB each.
Is my data safe?
Yes. Your API data is not stored or used for training, and all requests are sent over encrypted connections.
Other models
How to Get Started
Top Up Your Balance
Add credits via USDT or Epay. One balance works across all models and providers.
Top Up NowCreate Your API Key
Sign up and create an OpenAI-compatible API key in 3 seconds. No credit card required.
Create API KeyPlug Into Your Code or Agent
Swap in your APIMaster key and base URL — works with any OpenAI-compatible tool: Claude Code, Cursor, LangChain, LiteLLM, and more.
View Docs