HN Showcase
Back to browse
Nano Banana 2 – Sub-second AI image gen via Gemini 3.1 Flash

Nano Banana 2 – Sub-second AI image gen via Gemini 3.1 Flash

by naxtsass·Feb 27, 2026·1 point·0 comments
Visit ProjectView on HN

AI Analysis

●MidSlickEye Candy

Gemini 3.1 Flash wrapper with nice UX, but dozens of AI image frontends exist.

Strengths
  • •Custom streaming pipeline for preview tokens shows genuine optimization work
  • •Multi-region i18n via real-time search breakout data (automated rather than manual)
  • •Free tier removes friction and targets HN community directly
Weaknesses
  • •Completely dependent on Google's Gemini API—no differentiation if they launch their own UI
  • •'Speed' claim rests on Flash model, not architectural choice—any Gemini wrapper gets same latency
Category
AI/ML
Target Audience

Casual users wanting fast AI image generation; developers building image gen APIs

Similar To

Replicate · Hugging Face Spaces · midjourney.com

Post Description

Hi HN,

I wanted to see how far I could push the "Flash" in Gemini 3.1. Most AI interfaces feel bloated and slow, so I built Nano Banana 2 to focus on raw speed.

Technical Highlights:

Stack: Next.js + Edge Runtime to minimize TTFB.

Optimization: Custom streaming pipeline for Nano Banana 2 image preview tokens.

Locales: Automated i18n for 13 regions (MS, IL, PL, etc.) based on real-time search breakout data.

The goal was to build a globally responsive UI that feels native regardless of where the user is. Would love your feedback on the latency and the streaming implementation.

Free to try for the HN community: https://nano-banana2.me/

Similar Projects