Nano Banana

Google's image-edit model — the codename that became Gemini's flagship image stack

0Design & CreativeFreemium

Editorial Review

"Nano Banana" was Google's internal codename for its 2025 image model (Gemini 2.5 Flash Image) that originally shipped as Gemini's image-generation and image-edit capability. Google has since launched Nano Banana Pro (Gemini 3 Pro Image, November 2025) and Nano Banana 2 (Gemini 3.1 Flash Image, February 2026) as successors. The codename leaked to lmarena.ai's anonymous benchmark arena months before launch and dominated the leaderboard, generating viral discourse. By the time the product officially shipped under the Gemini brand, the "nano banana" name had stuck — and many buyers still search for it that way. Capability-wise it's a category leader on identity-preserving edits ("change the background but keep the person identical") and multi-image composition. Free access via the Gemini app makes it the cheapest serious AI image-edit tool for casual users; Gemini Advanced ($20/mo) and Vertex AI usage unlock higher limits and API access. Compared to Midjourney's stylized aesthetic and FLUX's photoreal naturalism, Nano Banana's strength is "follow my exact instruction on this exact image" — which is exactly what most working professionals need.

Key Features

Identity-preserving subject editsMulti-image composition (combine images)Long-prompt adherence

About Nano Banana

Nano Banana is the codename — and de facto product name — for Google's flagship image-edit and generation model, integrated into Gemini and accessible via the Gemini app, Vertex AI, and the Gemini API. Strong on identity-preserving edits, multi-image composition, and following long, specific prompts.