In short
- The totally free Gemini 2.5 Flash Image tool turns selfies into hyperrealistic 1/7-scale figurines, triggering explosive need.
- Users merely publish a full-body picture, paste an in-depth timely, and get toy-store-ready lead to seconds.
- Politicians and influencers are embracing the pattern, while innovative users fine-tune products, presents, and props for tailored dolls.
Google’s most current AI design has actually triggered another viral improvement pattern, with countless users turning themselves into mini Bandai-style figurines that look pulled from a Tokyo toy shop rack.
The tool, nicknamed Nano Banana after its internal codename at Google DeepMind, formally released as Gemini 2.5 Flash Image in late August. Within 2 weeks, users produced over 200 million images. The Philippines alone represents 25.5 million image productions simply days after its launch, making it the leading area worldwide for usage.
” From picture to figurine design in simply one timely,” Google’s Gemini group published on X September 1, beginning tutorials that spread out the pattern even more. Josh Woodward, Google’s VP, reported 10 million brand-new Gemini users signed up with particularly for the function.
Update: In the last 4 days, @GeminiApp has actually included:
+ 13M more newbie users (23M+ overall now)
+ 300M more images (500M+ overall now)@NanoBanana is unbelievable
— Josh Woodward (@joshwoodward) September 8, 2025
The digital figurines look like 1/7 scale antiques on clear acrylic bases, frequently showed beside product packaging boxes and computer system screens revealing 3D modeling software application. The AI catches facial functions, clothes information, and presents with sufficient precision that results often deceive audiences into believing they’re genuine item pictures.
This marks the 3rd significant AI picture pattern of 2025. ChatGPT’s GPT-4o triggered the Studio Ghibli wave in March, with users changing pictures into soft, anime-style pictures similar to Hayao Miyazaki’s movies. Sam Altman altered his X profile photo to a “Ghiblified” variation of himself as servers had a hard time under “scriptural need.”
April brought the Barbie Box Obstacle through ChatGPT, where individuals ended up being plastic dolls in blister product packaging, total with devices like small laptop computers and coffee mugs. LinkedIn experts welcomed it for individual branding, turning themselves into “executive action figures.”
Nano Banana isn’t the only gamer in this area. Alibaba’s Seedream4, launched simply a couple of days after Google’s design, produces outputs with equivalent quality and in some cases much better consistency in complicated presents. For those choosing open-source options, Flux Kontext provides the most effective alternative for regional implementation. It incorporates into complicated workflows and stays the only design without content constraints, though that versatility includes the normal obligations of self-hosted AI.
How to turn yourself into a figurine, free of charge
Developing your own Nano Banana figurine takes under a minute. Check out gemini.google.com or open the Gemini app– the fundamental variation is totally totally free, though your everyday generations are restricted.
Click the alternative to attempt Nano Banana, which you can discover it in the banner on top. It is likewise under “tools” as “Develop Images.” (Google is revealing banana emojis all over, so you will not have any difficulty discovering it.)
Next, publish your picture or the picture of the individual you wish to become a doll, ideally a full-body picture with great lighting. You can do that by clicking the “Plus” button beside “tools.” We trust you will be accountable, ok?
Once the image has actually been published, paste in this timely:
” Develop a 1/7 scale advertised figurine of the character in the photo, in a sensible design, in a genuine environment. The figurine is put on a computer system desk. The figurine has a round transparent acrylic base, without any text on the base. The material on the computer system screen is a 3D modeling procedure of this figurine. Beside the computer system screen is a toy product packaging box, developed in a design similar to premium collectible figures, printed with initial art work.”
There are some tweaks you can make to tailor your doll. Defining products like “PVC figure” or “refined resin” increases realism. Dynamic presents work much better than fixed standing positions– arms outstretched or mid-action yield more persuading figures.
If you do not believe the design is sensible enough, including something like “The face is precisely the exact same” or “the face stays unchanged,” suffices. However beware, you might wind up with something too sensible and not actually doll-like.
Do not go too insane, though. The totally free tier provides minimal everyday edits, while the professional variation eliminates constraints. Some platforms, like Freepik, offer users endless models (and endless Wan 2.2 videos too) with the tradeoff of a little deterioration in quality after a really generous limit is fulfilled.
Advanced users chain numerous triggers for complicated scenes. After creating the base figurine, you can enhance your doll with additional models– asking it, for instance, to alter the clothing, include props, change the present, and so on. Nano Banana’s impressive character recommendation makes it a really effective tool that will not deteriorate your face excessive after numerous models. This is something that previous designs utilized to fight with.
The figurine visual has actually shown especially sticky on social networks. Political leaders in India and the Philippines published their mini variations, while TikTokers utilize the hashtag #NanoBanana to share significantly sophisticated scenes with numerous figures and custom-made dioramas.
Typically Smart Newsletter
A weekly AI journey told by Gen, a generative AI design.