Nano Banana 2

According to the page, Nano Banana 2 is presented as Google's latest image model (Gemini 3.1 Flash Image), positioned as delivering Pro-level quality at Flash speed for creators, marketers, and developers. It supports text-to-image generation and image-to-image editing, uses Google Search and Image Search grounding to reference current visuals, renders text in images with about 92% character accuracy including multilingual text, offers four resolution tiers (512px, 1K, 2K, 4K) and 14 aspect ratios, and includes a Thinking Mode with Minimal, High, and Dynamic levels. All outputs carry a SynthID watermark, with C2PA Content Credentials also mentioned as being implemented. The page also lists related LoRA tools such as headshot, caricature, thumbnail, pet meme, pixel art, and sticker generators, and a comparison with Nano Banana Pro showing Pro takes 10–20 seconds per generation.

Key features

• Fast generation in about 4–6 seconds per request
• Subject consistency with up to 5 characters and 14 objects in one workflow
• Up to 14 reference image uploads
• Google Search and Image Search grounding
• Resolution tiers of 512px, 1K, 2K, and 4K (4096px)
• 14 aspect ratios
• Advanced text rendering (about 92% character accuracy)
• Translation of on-image text while preserving other visual elements
• Thinking Mode with Minimal, High, and Dynamic levels
• Image-to-image editing
• SynthID watermark on all outputs
• C2PA Content Credentials (stated as being implemented)

Best suited for

• Infographics and data visualization
• Multilingual marketing and localization
• Storyboarding and narratives
• Marketing mockups and packaging visuals
• Product showcases
• Rapid iteration and high-volume image generation

Pricing

Not stated on the website