Nano Banana 2Faster. Sharper. Smarter.
Think First. Generate Better.
Fast Generation. Full Fidelity.
Nano Banana 2 generates vibrant, high-fidelity images in 5-10 seconds. Built on Flash architecture, it reasons about composition and spatial relationships before rendering, but does it at speed. Rich color, punchy contrast, and coherent structure without the wait.
Typography That Actually Works
Render legible, correctly spelled text directly in generated images across multiple languages. Nano Banana 2 handles signage, labels, watermarks, and typographic compositions with character-by-character validation. From product packaging to social media graphics, the text comes out right the first time.
Edit With Words, Not Masks
Describe changes in plain language and Nano Banana 2 executes them. Replace backgrounds, change colors, add or remove objects, adjust lighting, all without masks or manual selection. Support for up to 14 reference images and multi-image compositing makes complex edits as simple as a sentence.
Generate and edit, one API
Text-to-image for new creations, image editing for refinements. Both endpoints share the same model and pricing.
See what Nano Banana 2 can create
Copy any prompt below and try it yourself in the playground.
"A hyperrealistic product photo of a matte ceramic coffee mug on a wooden table, morning light streaming through a window casting soft shadows, steam rising from the cup with the text 'GOOD MORNING' embossed on the side in a clean serif font, shallow depth of field, 85mm lens"
"A cinematic portrait of an elderly Japanese fisherman mending nets at dawn, weathered hands in sharp focus, fishing village harbor in soft bokeh behind him, golden hour side lighting, documentary photography style, Leica 50mm Summilux wide open"
"An aerial view of a bioluminescent bay at night, glowing blue-green plankton tracing the wake of a small wooden boat, stars reflected in the calm water, long exposure effect, the Milky Way arching overhead, National Geographic cover quality"
"A detailed architectural visualization of a modern hillside house with floor-to-ceiling glass walls, infinity pool merging with the ocean horizon, warm interior lighting at dusk, tropical vegetation, rendered in photorealistic style with accurate reflections and volumetric light"
A few lines of code.
Reasoning-powered output.
fal.ai handles the infrastructure: fast inference, auto-scaling, and a developer-friendly API. No GPUs to manage.
- Serverless: scales to zero, scales to millions
- Pay per image, no minimums
- Python and JavaScript SDKs, plus REST API
import fal_client
result = fal_client.subscribe(
"fal-ai/nano-banana-2",
arguments={
"prompt": "A ceramic mug on a wooden table,
morning light, 'GOOD MORNING' text",
"resolution": "2K",
}
)
# result["images"][0]["url"] → your generated imageCommon questions about Nano Banana 2
What is Nano Banana 2?
Nano Banana 2 is Google's next-generation image model built on the Gemini 3.1 Flash Image architecture. It combines reasoning-guided generation with speed and efficiency, producing high-fidelity images with accurate text rendering, character consistency, and natural-language editing, all in 5-10 seconds per generation.
What's new compared to Nano Banana 1?
Nano Banana 2 introduces reasoning-guided generation (the model plans before rendering), dramatically improved text rendering with character-by-character validation, native multi-resolution output (1K/2K/4K), character consistency for up to 5 people, multi-image compositing with up to 14 reference images, and optional web search grounding for real-world factuality. Generation speed is also significantly faster at 5-10 seconds per image.
What resolutions does Nano Banana 2 support?
Nano Banana 2 generates natively at 1K, 2K, and 4K resolution across a wide range of aspect ratios including 1:1, 16:9, 9:16, 21:9, 4:3, 3:2, and more. All resolutions are true native output, not upscaled.
How does image editing work?
The editing endpoint accepts one or more reference images along with natural language instructions. Describe changes like "replace the background with a beach sunset" or "change the shirt color to blue" without providing masks. Supports up to 14 reference images and 1-4 output variations per request.
How much does Nano Banana 2 cost on fal.ai?
Pricing is pay-per-image with no minimums or subscriptions. Check the model page for the latest pricing details.
Can Nano Banana 2 render text accurately in images?
Yes. Nano Banana 2 achieves state-of-the-art accuracy in text rendering with character-by-character validation, producing legible typography in multiple languages directly within generated images. This makes it well suited for product packaging, signage, social media graphics, infographics, and any visual that requires readable text.
How do I get started with the API?
Install the fal.ai SDK (Python or JavaScript), grab an API key from your dashboard, and make your first request in a few lines of code. The API is serverless, so there are no GPUs to manage and no infrastructure to set up. Check the API documentation for all available parameters.
Can I use Nano Banana 2 for commercial projects?
Yes. Images generated through the fal.ai API can be used in commercial projects. All outputs include SynthID digital watermarking for provenance. Check fal.ai's terms of service for full details on usage rights and licensing.
Ready to create?
Start generating stunning images with Nano Banana 2 on fal.ai.