Copy-paste prompts for portraits, characters, cinematic scenes, products, and more — tested on the #1-ranked image model on the planet.
By Duncan Rogoff
Based on 0 → 110K followers in under 12 months using AI-first content systems.
<aside> 🚀 Want the full AI content system behind this? The Build Room is a community where I share every workflow, prompt, and playbook I use to create daily.
👉 Join at skool.com/buildroom
</aside>
On February 26, 2026, Google launched Nano Banana 2 (Gemini 3.1 Flash Image). Within hours it hit #1 on the Artificial Analysis Image Arena — a blind human evaluation leaderboard — at roughly half the API price of its predecessor, Nano Banana Pro.
But the speed and price are the headline. Here is what actually matters for creators:
<aside> ⚡ 2–5x faster than Nano Banana Pro — same quality, less waiting.
</aside>
<aside> ✅ Readable text that actually works — headlines, infographics, and banners with sharp, legible copy.
</aside>
<aside> ✅ Character consistency across up to 5 characters — same face, same outfit, across multiple images in a sequence.
</aside>
<aside> ✅ 4K output — print-ready, broadcast-ready, detail that holds up at full resolution.
</aside>
<aside> ✅ Real-world knowledge via Google Search — NB2 understands current brands, events, and visual culture, not just training data from 2 years ago.
</aside>
The 20 prompts below are organized by use case. Each one comes with a note on why it works — because copy-paste without understanding is just luck.
NB2's portrait rendering is its most commercially useful capability. These three prompts cover influencer content, corporate headshots, and gaming creator content.