By News18,Sahas Mahapatra
Copyright news18
Months after the viral ChatGPT Ghibli trend, the Internet has found a new favourite way to generate pictures of themselves. The latest craze, the Nano Banana AI saree trend, has taken social media by storm, with numerous users, including celebrities, sharing their 3D-printed figurines created via Google Gemini 2.5 Flash Image.
In this trend, users can upload daily images and have them transformed into beautiful, Bollywood-style throwback portraits, featuring them in a flowing saree with flowers adorning their hair. However, there are a few important points to consider before creating your AI-generated pictures.
After several users reported that the Google Gemini Nano Banana AI saree trend distorted their faces and produced mismatched images due to unclear or incorrect prompts, experts recommend using precise and appropriate inputs to generate more realistic graphics.
Tips And Common Mistakes To Avoid For Perfect AI-Generated Nano Banana Saree Images
Using Group or Low-Quality Pictures
The AI model often struggles to process crowded images. For the best results, upload sharp, single photographs with a clear and visible face.
Wrong Prompts
To get the best results, the user must provide clear prompts on the AI platform. Instructions such as “make me look like a Bollywood actress” would not produce the intended outcomes. So, be specific: describe the saree colour, fabric, lighting style, and background.
Don’t Ignore The Background
Bollywood aesthetics rely on stunning backdrops. If you don’t provide one, the AI may select plain or mismatched scenes.
Avoid Overloading
Don’t attach too many details to your prompt, as it might confuse the AI, and you wouldn’t get the desired results. So stick to 3–4 concise visual cues instead of long paragraphs.
Facial Consistency Is Important
To avoid having your face drastically altered, fill in the prompt with lines like ‘keep the same facial features’ and ‘do not change the face shape.’
Nano Banana is Google DeepMind’s latest AI-powered image generation and editing model, released in August 2025.
Recently, Google updated its policy on image generation limits. Previously, free users could generate up to 100 images per day, while Pro and Ultra subscribers had a daily limit of 1,000. However, due to a significant increase in usage, these fixed limits were quietly removed without much public announcement, reflecting evolving user needs, platform development and the challenges of scaling the technology.
