Nano Banana
Google's conversational image editor: natural-language edits with character consistency.
About
Nano Banana is the public name for Google's Gemini 2.5 Flash Image model, launched August 26, 2025. It generates and edits images through natural-language conversation: upload a photo and ask to change the background, restyle a room, or merge multiple images, then iterate with follow-up prompts. Its standout trait is consistency — the same person, pet, or product stays recognizable across edits and scenes, which made it the top-rated image-editing model on LMArena at launch.
It is not a replacement for precision design tools. Output caps at 1024px-class resolution, every image carries a visible AI watermark plus invisible SynthID tagging in the Gemini app, and complex multi-element compositions can still drift. Google now points new workloads to the Nano Banana 2 family for higher resolution.
What it can do
- Natural-language photo editing
- Multi-image fusion and style transfer
- Character and object consistency across edits
- Conversational iterative refinement
- Text-to-image generation
Setup
No install. Use it in the Gemini app (gemini.google.com), Google AI Studio, or via the Gemini API with responseModalities set to TEXT and IMAGE. API pricing is per output token (~$0.039 per 1024px image at launch pricing).
Prerequisites
A Google account for the app; a Gemini API key from Google AI Studio for programmatic use.
Use cases
Works with
Trust score — 9/10
Scored 9/10 (verified tier, metrics-v1). Evidence: Verified via vendor site/docs/GitHub and reputable coverage, wave 44 (tools-b); Trending score 92/100 from researcher assessment.
Sources
Every fact above was checked against these sources. Where a claim comes from the provider and couldn't be independently verified, the text says so.