← Models

Nano Banana Pro

Google · Nano Banana

Google's stable Gemini 3 Pro Image model for professional visual assets and grounded image generation.

Type
image
Context
N/A
Max Output
N/A
Status
current
API Access
Yes
License
proprietary
image-generation image-editing google creative visual stable grounding thinking
Released May 2026 · Updated June 8, 2026

Overview

Freshness note: Image-model capabilities, safety behavior, lifecycle dates, and pricing can change quickly. This profile is a point-in-time snapshot last verified on June 8, 2026.

Nano Banana Pro is Google’s professional image-generation and editing route in the Gemini stack. The current stable Gemini API model ID is gemini-3-pro-image; Google’s lifecycle notes listed the older gemini-3-pro-image-preview route for shutdown on June 25, 2026, so new work should use the current stable ID.

The model is the higher-quality counterpart to Nano Banana 2. Use Nano Banana Pro when the task needs professional asset production, grounded visual reasoning, complex layouts, or stronger text rendering; use Nano Banana 2 when speed and cost matter more.

Capabilities

Nano Banana Pro is designed for complex visual instructions, higher-fidelity text rendering, localized editing, and professional creative controls. Google’s current model page describes it as a reasoning-driven engine for professional-grade image editing and generation, high-fidelity product mockups, complex graphic design, and factual data visualizations that need accurate text and real-world grounding.

The API docs also list search grounding, structured outputs, thinking, Flex inference, Priority inference, Batch API, and image generation support. That makes it especially useful for visual explainers, timely infographics, product visuals, and assets that need factual context rather than only style transfer.

Technical Details

This is an image-native model, so Signal Lens stores contextWindow and maxOutput as 0 and treats token-style fields as N/A for UI purposes.

Current public anchors:

  • API model ID: gemini-3-pro-image.
  • Inputs: image and text.
  • Outputs: image and text.
  • Input token limit: 65,536.
  • Output token limit: 32,768.
  • Supported capabilities: Batch API, Flex inference, Priority inference, image generation, search grounding, structured outputs, and thinking.
  • Not supported: caching, code execution, file search, function calling, live API, audio generation, URL context, and Google Maps grounding.
  • Image-generation guide support for 1K, 2K, and 4K outputs.

Pricing & Access

Google exposes Nano Banana Pro through Gemini product surfaces, Google AI Studio, and the Gemini API. Current standard Gemini API pricing lists:

  • Input: 2per1Mtextorimagetokens,withimageinputshownas560tokensorabout2 per 1M text or image tokens, with image input shown as 560 tokens or about 0.0011 per image.
  • Output text and thinking: $12 per 1M tokens.
  • Output images: 120per1Mimage−outputtokens,equivalentto120 per 1M image-output tokens, equivalent to 0.134 for 1K/2K images and $0.24 for 4K images.
  • Batch and Flex: 1per1Minputtexttokens,1 per 1M input text tokens, 0.0006 per input image, 6per1Moutputtext/thinkingtokens,andequivalentoutputimagepricesof6 per 1M output text/thinking tokens, and equivalent output image prices of 0.067 for 1K/2K and $0.12 for 4K.
  • Priority: 3.60per1Minputtokens,3.60 per 1M input tokens, 21.60 per 1M output text/thinking tokens, and $216 per 1M image-output tokens.

Search grounding has a shared Gemini 3 allowance of 5,000 prompts per month, then $14 per 1,000 search queries.

Best Use Cases

Use Nano Banana Pro for brand mockups, product visuals, visual explainers, diagram-style assets, multilingual text-in-image work, and grounded infographics where the output needs to look closer to a finished creative asset than a rough sketch.

Use Nano Banana 2 instead when speed, cost, and high-volume iteration matter more than maximum text fidelity or professional layout control.

Comparisons

  • Nano Banana 2 (Google): Faster Gemini 3.1 Flash Image route for high-volume iteration; Pro is the higher-quality Gemini 3 Pro Image route.
  • Imagen 4 (Google): Dedicated Imagen family now scheduled for endpoint shutdown; Nano Banana Pro is the safer professional Google image route.
  • GPT Image 2 (OpenAI): Strong OpenAI-native creative model; Nano Banana Pro is more attractive inside Google AI Studio, Gemini, Ads, and Search-grounded workflows.