Nano Banana 2

Overview

Freshness note: Model capabilities, limits, lifecycle dates, and pricing can change quickly. This profile is a point-in-time snapshot last verified on July 6, 2026.

Nano Banana 2 is Google’s stable Gemini 3.1 Flash Image model for efficient image generation and conversational image editing. The current Gemini API model ID is gemini-3.1-flash-image; Google’s lifecycle notes listed the older gemini-3.1-flash-image-preview route for shutdown on June 25, 2026, so new work should use the current stable ID.

Google now presents Nano Banana as a four-route family: Nano Banana 2 Lite for lowest latency and cost, Nano Banana 2 as the general workhorse, Nano Banana Pro for professional assets, and legacy Gemini 2.5 Flash Image for older integrations.

Capabilities

Google positions Nano Banana 2 as the high-efficiency workhorse between Lite and Pro: faster and cheaper than Pro, but more capable than Lite for richer editing and higher-resolution output. Current docs highlight 0.5K, 1K, 2K, and 4K output options, Image Search Grounding, additional extreme aspect ratios, improved aspect ratio adherence, better image quality and consistency, and improved international text rendering.

Those improvements make the model more useful for infographics, product explainers, localized visuals, diagram generation, and structured visual storytelling, not just aesthetic prompt art.

Technical Details

As an image-generation model, Signal Lens stores contextWindow and maxOutput as 0 and treats them as N/A in the UI. Google’s current API card lists:

• Model ID: gemini-3.1-flash-image
• Inputs: text and image/PDF
• Outputs: image and text
• Input token limit: 131,072
• Output token limit: 32,768
• Supported capabilities: Batch API, image generation, search grounding, and thinking
• Not supported: caching, code execution, file search, function calling, live API, structured outputs, URL context, audio generation, and Google Maps grounding

Google’s image generation guide lists 0.5K, 1K, 2K, and 4K output tables for Gemini 3.1 Flash Image, including aspect ratios from 1:8 through 8:1.

Pricing & Access

Google’s current Gemini API pricing lists the standard Nano Banana 2 route at:

• Input: $0.50 per 1M text or image tokens
• Output text and thinking: $3 per 1M tokens
• Output images: $60 per 1M image-output tokens
• Equivalent image prices: $0.045 for 0.5K,$0.067 for 1K, $0.101 for 2K, and$0.151 for 4K

Batch pricing is lower: $0.25 per 1M input tokens,$1.50 per 1M output text/thinking tokens, and $30 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 2 for infographics, product explainers, localized image variants, diagram generation, and high-speed visual iteration where good instruction following, cost, and latency matter more than maximum professional layout control. Use Nano Banana 2 Lite when 1K output and low cost are enough; stay on Nano Banana 2 when grounding, 2K/4K options, or stronger multi-turn image work matter.

Comparisons

• Nano Banana 2 Lite (Google): Lower-cost 1K route for high-volume simple image generation and edits.
• Nano Banana Pro (Google): Higher-fidelity Gemini 3 Pro Image route for professional assets; Nano Banana 2 is the faster and cheaper iteration lane.
• Imagen 4 (Google): Dedicated Imagen route now scheduled for endpoint shutdown; Nano Banana 2 is the safer default for new Google image work.
• Midjourney-style tools: Better for creator-led exploration, while Nano Banana 2 is more relevant to structured product and workflow integration.