LLM Reference

Nano Banana Pro (Gemini 3 Pro Image)

Researched today

Last refreshed 2026-05-31. Next refresh: weekly.

ProprietaryMultimodalAgentsVisionJSON / Tool use

Nano Banana Pro (Gemini 3 Pro Image) is worth evaluating for agents, vision, and json / tool use when its provider route and context window match the workload.

Decision context: Agents task fit, 1 tracked provider route, and research from 2026-05-31.

Use it for

  • Teams evaluating agents, vision, and json / tool use
  • Workloads that can use a 66k context window
  • Buyers comparing 1 tracked provider route

Do not use it for

  • Workloads where another current model has stronger sourced task evidence

Cheapest output

$120.00

Google AI Studio per 1M tokens

Provider routes

1

Tracked API hosts

Quality / dollar

Unknown

No task benchmark coverage yet

Freshness

2026-05-31

Researched today

fresh

Top use-case fit

Agents

Included by capability and metadata signals in the decision map.

Vision

Included by capability and metadata signals in the decision map.

JSON / Tool use

Included by capability and metadata signals in the decision map.

Provider price ladder

ProviderInput / 1MOutput / 1MBatch in / outRoute
Google AI Studio$2.00$120.00$1.00 / -
Serverless

Benchmark peer barsfor Agents

No task-mapped benchmark peers are available for this model yet.

Migration checks

No linked migration route is available for this model yet.

About

Nano Banana Pro is the GA Gemini 3 Pro Image model for high-fidelity image creation through the Gemini API. It is aimed at complex graphic design, product mockups, data visualizations, and accurate text rendering, and replaces gemini-3-pro-image-preview retiring on 2026-06-25.

Nano Banana Pro (Gemini 3 Pro Image) has a 66k-token context window.

Nano Banana Pro (Gemini 3 Pro Image) input tokens at $2/1M, output at $120/1M.

Capabilities

VisionMultimodalReasoningTool UseStructured OutputsBatch API

API Versions

gemini-3-pro-image

Rankings

Specifications

FamilyGemini 3
Released2026-05-28
Context66k
Max output32,768
ArchitectureDecoder Only
Knowledge cutoff2025-01
Specializationimage-generation
LicenseProprietary
Trainingpretrained

Created by

Pioneering artificial intelligence research.

London, United Kingdom
Founded 2014
Website

Providers(1)