{"id":110504,"date":"2025-09-01T09:37:17","date_gmt":"2025-09-01T09:37:17","guid":{"rendered":"https:\/\/www.newsbeep.com\/au\/110504\/"},"modified":"2025-09-01T09:37:17","modified_gmt":"2025-09-01T09:37:17","slug":"nano-banana-the-ai-thats-ripe-for-image-editing","status":"publish","type":"post","link":"https:\/\/www.newsbeep.com\/au\/110504\/","title":{"rendered":"Nano Banana: The AI that\u2019s ripe for image editing"},"content":{"rendered":"

Two weeks ago, a mysterious model called \u201cNano Banana\u201d started generating a buzz on LMArena, a blind-testing platform where anonymous AI models are pitted against each other. Users on Reddit forums dedicated to AI imagery were frothing over its capabilities, particularly for tasks involving image editing.\u00a0<\/p>\n

\"\"<\/p>\n

Different flavours of Shaun Davies (Gemini)<\/p>\n

The smart money was on a stealth project from Google, and these rumours turned out to be true. \u201cNano Banana\u201d was officially unveiled last week as Google\u2019s Gemini 2.5 Flash Image model, and it\u2019s now widely available in Gemini and Google AI Studio. While it\u2019s early days, this feels like a significant model update of great interest to marketers, creatives and anyone who\u2019s struggled with getting consistent and usable image outputs from AI.\u00a0<\/p>\n

What is Nano Banana?<\/p>\n

The primary claim of Google\u2019s new top banana is that it solves one of generative AI\u2019s most persistent problems: image consistency.<\/p>\n

ADVERTISEMENT<\/p>\n

Up until now, image models have not been good at keeping the details in photos the same from one generation to the next. Let\u2019s say you upload an image of yourself as a seed for a prompt. The first image the model spits out will likely have made a few noticeable changes to your appearance. The next image will mutate a little more, and before you know it you\u2019re looking at a freakish, fun-house mirror of yourself.<\/p>\n

Google\u2019s claim is that Nano Banana goes a long way to fixing that, with three key improvements.<\/p>\n

Character and Style Consistency: The model can maintain the appearance of a person, object, or artistic style across a whole series of prompts and edits.
\nPrecise, Conversational Editing: Often called \u201cmulti-turn editing\u201d, this allows you to have a back-and-forth conversation to refine an image. You can make specific, targeted changes (like blurring a background or altering a pose) using natural language, without the model losing track of the original request.
\nMulti-image Fusion: The model can blend elements from multiple source images into a single, cohesive visual.<\/p>\n

Do these claims survive under scrutiny? Or is this just another round of hype from a tech company desperate to get you using their products? For the rest of this article, we\u2019ll put Nano Banana under pressure to find out if it\u2019s underripe, rotten to the core, or perfectly ripe for the picking.<\/p>\n

Test One: Headshots, mohawks and consistency<\/p>\n

Recently I sat for a photo shoot with the brilliant photographer Neil Bennett and I was extremely happy with the result. But looking at this carefully constructed version of myself, I can\u2019t help but think back to my younger days when I would cringingly declare that I was too \u201cpunk rock\u201d for such corporate malarkey. So I decided to transform one of my portraits in a few different ways to see how well Nano Banana\u2019s claims of consistency stood up.<\/p>\n

\"\"<\/p>\n

Shaun Davies (Neil Bennett)<\/p>\n

To make it a fair fight, I ran the exact same experiment on GPT-5. The goal was simple: could they make my headshot punk rock, while still remembering my actual face?<\/p>\n

My first prompt was simple. I am bald. So I asked the models: \u201cGive this man a full and lustrous head of hair.\u201d<\/p>\n

Both models handled this reasonably well, though GPT-5\u2019s interpretation already began to stray, giving me a slightly fuller face and a different hair texture.\u00a0<\/p>\n

\"\"<\/p>\n

\u201cHair\u201d by ChatGPT<\/p>\n

Nano Banana\u2019s version felt more like a plausible version of me, although the hair itself looked like a wig and it introduced some strange artefacts into my blue suit.\u00a0<\/p>\n

\"\"<\/p>\n

\u201cHair\u201d by Gemini<\/p>\n

One thing I immediately noticed was latency. GPT-5 took about 80 seconds to output its image, while Gemini 2.5 Flash was done within 30.<\/p>\n

The next prompt upped the ante on the punk thing.<\/p>\n

Make the hair a mohawk dyed in multiple colours (red, black, purple). Put me in a leather jacket and torn T-shirt.<\/p>\n

This is where the wheels fell off for one of our contenders. Nano Banana diligently added the mohawk and leather jacket, keeping my facial structure intact.\u00a0<\/p>\n

\"\"<\/p>\n

\u201cMohawk\u201d by Gemini<\/p>\n

But GPT lost the plot and generated someone else entirely.<\/p>\n

\"\"<\/p>\n

\u201cMohawk\u201d by ChatGPT<\/p>\n

The final prompt was : \u201cMake me thinner and 20 years younger.\u201d<\/p>\n

Nano Banana processed the request and delivered this:<\/p>\n

\"\"<\/p>\n

\u201cThin young mohawk\u201d by Gemini<\/p>\n

It\u2019s pretty good. It more or less looks like me in my late 20s, if I\u2019d ever had the guts to grow a mohawk.\u00a0<\/p>\n

GPT-5, on the other hand, went totally off the rails:<\/p>\n

\"\"<\/p>\n

\u201cThin young mohawk\u201d by ChatGPT<\/p>\n

A fine-looking young punk, to be sure, but most definitely not me. He looks more like my long-lost nephew who plays bass in a band called Societal Collapse or Plegm or something. For consistency across multiple generations, Nano Banana was in a different league.<\/p>\n

Test 2: Conversational editing<\/p>\n

Of course, a tool\u2019s ability to remember a face through a punk rock makeover is one thing. But how does it handle inserting an entirely new object into a picture, while maintaining a consistent landscape? For the second test, I used a photo of the sky at sunset\u2014one of those moments that feels profound in person but looks like a dull smudge on your phone\u2014and gave both models a two-part challenge.<\/p>\n

First, the prompt:<\/p>\n

Change this scene so that instead of the cloud, there is a gigantic space monster that looks like an interstellar version of a lion\u2019s mane jellyfish hovering in the skyline\u2026 The effect should be realistic, slice of life, camera phone.<\/p>\n

This round went to GPT-5. Its jellyfish had a more cinematic, almost ominous quality, blending into the hazy atmosphere in a way that felt genuinely surreal.\u00a0<\/p>\n

\"\"<\/p>\n

\u201cJellyfish\u201d by ChatGPT<\/p>\n

Nano Banana\u2019s version was more vibrant and ethereal, but was very much a jellyfish, a beautiful but less subtle take on the brief.<\/p>\n

\"\"<\/p>\n

\u201cJellyfish\u201d by Gemini<\/p>\n

For the next prompt, I decided to see the models handled a physics-defying concept.<\/p>\n

Make the jellyfish into an upside down bowl of ramen noodles. Literally make the tentacles into noodles that are yellow, thick and curly.<\/p>\n

I preferred Nano Banana\u2019s version here, a chaotic tumble of noodles cascading from a porcelain bowl.\u00a0<\/p>\n

\"\"<\/p>\n

\u201cRamen\u201d by Gemini<\/p>\n

GPT-5\u2019s attempt was tamer, its noodles more like decorative squiggles under a plain brown dome. Honestly, though, there wasn\u2019t a lot in it.<\/p>\n

\"\"<\/p>\n

\u201cRamen\u201d by ChatGPT<\/p>\n

But here\u2019s the observation that matters more than the aesthetics of alien jellyfish or flying pasta. While GPT-5 subtly altered the landscape with each generation, Nano Banana\u2019s consistency was flawless. Every house, tree, and distant building in the background remained identical across the edits. It didn\u2019t just add the requested object; it did so while preserving the integrity of the original image.<\/p>\n

This should grab the attention of any marketing or creative professional. The ability to lock down a background and make specific, iterative changes is a genuine step forward. Imagine mocking up a product in a dozen different real-world settings without the AI deciding to creatively reinterpret the footpath or remove a competitor\u2019s store from the background. It\u2019s a level of control that could move these tools from the realm of amusing toys to practical, reliable workhorses.<\/p>\n

Test 3: Xenomorph ski buddy<\/p>\n

The final test was for \u201cmulti-image fusion,\u201d another of Nano Banana\u2019s key features. This allows the model to blend elements from two separate images. Having recently been skiing and also watching the new Alien: Earth series, I thought it could be fun to create a xenomorph ski buddy. I fed both models a photo of me on the slopes and an image of H.R. Giger\u2019s finest monster.<\/p>\n

The prompt was specific:<\/p>\n

Extend the picture of the man in the snow. He is skiing with his best friend, an alien. Insert the alien into the photo with the man. They are both smiling and happy. They have their arms\/claws around each other.<\/p>\n

ChatGPT produced a far better-composed image. Its xenomorph was more dynamic, the framing was more interesting, and the overall scene felt more like a photograph. But what the hell did it do to my face? It looks like a rubber Halloween mask.<\/p>\n

\"\"<\/p>\n

\u201cBuddies\u201d by ChatGPT<\/p>\n

Nano Banana\u2019s composition was static, there was an extra pair of skis and the xenomorph looked more like a statue I was awkwardly posing with. But the \u201cme\u201d in the photo was still me. It had faithfully preserved my likeness from the source image.<\/p>\n

\"\"<\/p>\n

\u201cBuddies\u201d by Gemini<\/p>\n

A follow-up prompt to \u201cPut the alien in retro ski gear\u201d resulted in some truly magnificent 80s ski fashion. GPT produced a hilarious xenomorph with a huge grin, but it mangled my face further, giving me a weird elongated chin and plastic skin, like the star of an Aphex Twin video clip.\u00a0<\/p>\n

\"\"<\/p>\n

\u201cRetro buddies\u201d by ChatGPT<\/p>\n

Nano Banana diligently kept my image as is, but continued with the awkward staging.<\/p>\n

\"\"<\/p>\n

Reliability versus artistic flair<\/p>\n

There\u2019s a clear contrast between these models. Nano Banana is the dependable, if slightly literal, creative partner. It doesn\u2019t always have the most artistic flair, but it listens, it remembers, and it doesn\u2019t go rogue. In every test, Nano Banana prioritised faithfulness to the source material. My face remained my face; my balcony view remained my balcony view.<\/p>\n

Professionals don\u2019t always need a tool to have a dazzling vision of its own; often it\u2019s more important that the idea is executed precisely and consistently.<\/p>\n

On the other hand, GPT-5\u2019s image generation tends to produce more imaginative and aesthetically pleasing results. But it is impossible to keep details consistent across images for more than one generation.<\/p>\n

Bonus: Vibe coding with Nano Banana<\/p>\n

Recognising that not everyone wants to engage in a multi-turn conversation to get a result, Google has also packaged this technology into a series of pre-built apps within AI Studio. There are tools for retouching photos and one particularly neat app called Past Forward<\/a> that can place your photo into different decades, complete with era-appropriate fashion and film stock effects. The image at the top of this article is a composite of Past Forward outputs using my headshot.\u00a0<\/p>\n

These apps make the power of the model accessible for specific, common tasks. But using AI Studio\u2019s Build function, you can also alter these apps to suit your own purposes, or even vibe code a whole new app with a simple series of text prompts. I will add a note of caution \u2013 vibe coding without engineering skills means you can\u2019t tell if your code is inefficient or has massive security loopholes in it. But it sure is fun to play around with.<\/p>\n

Ultimately, the choice of AI tool will always come down to the job at hand. If you\u2019re in the blue-sky phase of a campaign, brainstorming wild and unexpected concepts, a more creatively unpredictable model might be your best bet. But when it\u2019s time to execute\u2014to place a product, maintain a brand aesthetic, or ensure a CEO\u2019s headshot doesn\u2019t suddenly sprout a stranger\u2019s face\u2014reliability is paramount.<\/p>\n

It\u2019s well-priced too, at about 3 cents per image via the API, which opens up the possibility of producing hundreds of variants, quickly and at low cost, while maintaining that same consistency. Nano Banana, with its steadfast memory and literal interpretation, is shaping up to be useful and, most importantly, predictable.<\/p>\n

Shaun Davies is the founder and principal of The AI Training Company<\/a>.<\/p>\n","protected":false},"excerpt":{"rendered":"Two weeks ago, a mysterious model called \u201cNano Banana\u201d started generating a buzz on LMArena, a blind-testing platform…\n","protected":false},"author":2,"featured_media":110505,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[20],"tags":[256,254,255,64,63,105],"class_list":{"0":"post-110504","1":"post","2":"type-post","3":"status-publish","4":"format-standard","5":"has-post-thumbnail","7":"category-artificial-intelligence","8":"tag-ai","9":"tag-artificial-intelligence","10":"tag-artificialintelligence","11":"tag-au","12":"tag-australia","13":"tag-technology"},"_links":{"self":[{"href":"https:\/\/www.newsbeep.com\/au\/wp-json\/wp\/v2\/posts\/110504","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.newsbeep.com\/au\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.newsbeep.com\/au\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/au\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/au\/wp-json\/wp\/v2\/comments?post=110504"}],"version-history":[{"count":0,"href":"https:\/\/www.newsbeep.com\/au\/wp-json\/wp\/v2\/posts\/110504\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/au\/wp-json\/wp\/v2\/media\/110505"}],"wp:attachment":[{"href":"https:\/\/www.newsbeep.com\/au\/wp-json\/wp\/v2\/media?parent=110504"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.newsbeep.com\/au\/wp-json\/wp\/v2\/categories?post=110504"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.newsbeep.com\/au\/wp-json\/wp\/v2\/tags?post=110504"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}