MODEL LINEAGE
中文
Model index
Google DeepMind·Nano Banana

Nano Banana 2

Brings 0.5K–4K generation, thinking, and image-search grounding at Flash speed and price.

CurrentImage generationReasoningMultimodalWeights not released
MODEL NUMBER207
First public date
2026-02-26
Organization
Google DeepMind
Family
Nano Banana
01

Specifications

Only published specifications are shown.

Maximum input
131,072 tokens↗
Maximum output
32,768 tokens↗
Input modalities
Text · Image
Output modalities
Image · Text
Open weights
No
Status
Current
02

Overview

Release dates

— Nano Banana 2 entered the Gemini API. ↗

03

Specifications alongside predecessor

Compared with Nano Banana ↗

PreviousNano Banana— · — ctx
CurrentNano Banana 2— · — ctx
04

Architecture & capabilities

Model capabilities

Conversational editingText renderingHigh resolution

Connected tools

Image search
05

Release history

API launch

Nano Banana 2 entered the Gemini API.

View source
06

Related models

Select a card to view the related model.

PredecessorNano BananaView details →SiblingNano Banana 2 LiteView details →SuccessorNano Banana 2.1View details →
07

Sources

01
Build with Nano Banana 2Google · Published 2026-02-26
Official · Verified02
Gemini 3.1 Flash Image model limitsGoogle · Checked 2026-09-07
Official · Verified
Suggest a correction

Copy this template, add the field, proposed value, and primary source, and share it with the archive maintainer.