zho-zho-zho/zho-nano-banana-creation — explained in plain English
Analysis updated 2026-07-03
Copy prompts from the collection to convert a portrait photo into a cartoon, doll, or Funko Pop style on Nano-Banana
Use the included prompts to generate product advertisement video clips that feature a specific person holding an item
Browse techniques for face blending, virtual makeup try-ons, and high-resolution image enhancement to find a workflow for your project
| zho-zho-zho/zho-nano-banana-creation | bytelegend/bytelegend | dingjikerbo/android-bluetoothkit | |
|---|---|---|---|
| Stars | 3,701 | 3,701 | 3,701 |
| Language | — | Java | Java |
| Setup difficulty | easy | moderate | moderate |
| Complexity | 1/5 | 3/5 | 3/5 |
| Audience | designer | developer | developer |
Figures from each repo's GitHub metadata at analysis time.
ZHO-nano-banana-Creation is a collection of AI image and video generation techniques shared by a creator who goes by ZHO. The repository documents 46 different creative workflows built around a tool called Nano-Banana, an AI-powered image and video generation platform. Most entries pair a short description of the effect with the exact prompt text used to achieve it, along with a link to the original post on X (formerly Twitter) where the result was shared. The repository is described as continuously updated. The techniques span a wide range of visual tasks. Some convert photographs into stylized outputs: turning a photo of a person into a collectible figurine with packaging, converting an architecture photo into a physical scale model render, or transforming a portrait into cartoon, doll, or Funko Pop styles. Others work with video, generating short clips that preserve a specific person's facial features while changing the camera angle, or producing product advertisement clips that place a specified person holding a product. There are also techniques for high-resolution image enhancement, object compositing, virtual makeup try-ons, lighting reference generation, and blending two faces to generate a child's appearance. Alongside the prompt collection, the repository links to three open-source apps the same author built, each offering a different interface for working with these techniques. Nano Bananary is a desktop window-style application. BananaPod is a canvas and whiteboard-style app with support for hand-drawn inputs and multi-frame selection. BananaFlow is a node-based workflow editor that combines Nano-Banana with video generation. The content is written primarily in Chinese, with some prompts provided in English. There is no code in this repository, it is a reference and inspiration document for people using AI generation tools.
A reference collection of 46 AI image and video generation techniques for the Nano-Banana platform, pairing exact prompts with example results for effects like style transfer, figurine creation, and product ad clips.
Setup difficulty is rated easy, with roughly 5min to a first successful run.
Mainly designer.
This repo across BitVibe Labs
Verify against the repo before relying on details.