N

Nano Banana

Description

The name Google uses for Gemini's built-in image generation and editing. It is a family of image models, not a separate website: you use it inside the Gemini app and chatbot (see the Gemini page), and developers can call it through the Gemini API; the same models are also offered in Google Cloud. You describe a picture in a conversation, or upload photos and ask to change them, and keep refining the result with follow-up messages. It accepts text, images and video as input and can combine up to 14 reference images. Output can reach 4K resolution, depending on the model. Every generated image carries Google's SynthID watermark. The family includes a premium model for complex tasks and a fast, cheaper lite model; older models stay available through the API.

Pros and cons

Pros

Image generation is available in the Gemini app without a paid plan
Edit your own photos by chatting, with up to 14 reference images
Every image carries a SynthID watermark that marks it as AI-generated

Cons

The most advanced model in the Gemini app needs a Google AI plan
The API has no free tier
It is not a standalone site: you reach it through Gemini, the API or Google Cloud

Key facts

Free access: Yes In the Gemini app image generation is available without a Google AI plan, with limits that Google may change; the most advanced model needs a plan, and the API has no free tier
Pricing: Pricing on the official site →
Developer: Google
Interface languages: English, Russian
Checked on: Oct 27, 2026

Editorial score

How strongly the AI Expertise editors recommend this tool for each of its categories right now, on a 0–10 scale. The score combines output quality, features, beginner friendliness and freshness.

Image Generation

8.7 / 10

Output quality 9.4

Low High

Features 7.0

Basic Rich

Beginner friendliness 8.5

Hard Easy

Freshness 10.0

Outdated Fresh

News about Nano Banana

06 Oct 2026

Google launches Nano Banana 2.1, a faster, cheaper image model replacing the February 2026 version. Prices halved: 1K images at 3.36¢, 4K at 7.56¢. Built on Gemini 3.6 Flash, it improves visual quality, text rendering, character consistency, panoramas, and infographics. Handles 14 reference images, maintains 4 characters/10 objects, connects to Search, offers thinking levels. Benchmarks show it sometimes beats Pro, but Pro still wins in practice for realism. Rolling out across Gemini app, Search, AI Studio, Flow, Stitch, Ads, Enterprise. Old model deprecated Oct 29. #GoogleAI #NanoBanana #ImageGeneration #Gemini #AIArt
05 Aug 2026

Google launches two new AI models: Nano Banana 2 Lite generates images in 4 seconds at $0.034/image (1K resolution), while Gemini Omni Flash creates 10-second videos via API at $0.10/second. Both available in Google AI Studio, Gemini API, and Enterprise Agent Platform. Google recommends chaining them — generate images with Nano Banana 2 Lite, then animate with Gemini Omni Flash. Models use SynthID watermarks for AI content verification. #GoogleAI #NanoBanana #GeminiOmniFlash #GenerativeAI #AIVideoGeneration
15 Jul 2026

Google launches Nano Banana 2 Lite for ultra-fast AI images (4 sec, $0.034/img) and Gemini Omni Flash for video generation via API (10-sec clips, $0.10/sec). Both models integrate via the Interactions API, enabling seamless image-to-video workflows. Nano Banana 2 Lite replaces the original Nano Banana, offering reliable prompt following and text rendering. Gemini Omni Flash supports conversational video editing with multimodal inputs. Google provides demo apps showcasing chained workflows: selfie-to-landmark animation, room-to-video design, and product-to-ecommerce video. Both use SynthID watermarks. Available now in Google AI Studio, Gemini API, and Enterprise Agent Platform. #GoogleAI #NanoBanana #GeminiOmniFlash #GenerativeAI #AItools