After seeing all of the banana-fanfare for Google’s latest generative AI instrument, I knew I needed to take it for a spin. Named Gemini 2.5 Flash Image, the mannequin upgrades your potential to edit your images natively in Gemini. AI lovers have referred to it because the “nano bananas” mannequin, spurred on by a sequence of banana-themed teasers from Google execs. In the few weeks it has been out, folks have created over 200 million AI photographs, and over 10 million folks have signed up to make use of the Gemini app, based on Josh Woodward, Google’s vice chairman of Google Labs and Gemini. Google has invested closely in its generative media fashions this yr, dropping up to date variations of its picture and video generator fashions at its annual I/O builders convention. Google’s AI video generator Veo 3 surprised with synchronized audio, a primary among the many AI giants. And creators have made greater than 100 million AI movies with Google’s AI filmmaker instrument, Flow. I’ve spent a number of time testing AI artistic software program, and I used to be excited to see what Google had cooked up. But my testing of two.5 Flash Image confirmed that simply because one thing has a flashy entrance doesn’t suggest it will at all times lives as much as its hype. Here’s how my expertise with Gemini nano bananas went: the nice, the unhealthy and the irritating.What laboredThe Gemini bananas mannequin is spookily good at including parts to present photographs, mixing AI-generated parts nicely into any image you snapped. It additionally maintains a decently secure degree of character consistency — which means the folks in my images weren’t too distorted or wonky after going by means of the AI processing. Those are each essential distinctions for AI picture applications, and one thing Google stated it had labored to enhance.You can see each of those traits on this image of my sister and me. Our common appearances are unchanged within the edited model (proper), exhibiting off that character consistency. I requested Gemini so as to add a 3rd sister who seemed much like the 2 of us, which it did scarily nicely by including a 3rd lady in between the 2 of us. Welcome to the household, scarily accurate-looking AI sister. Katelyn Chedraoui/GeminiI used to be additionally fairly impressed with how shortly Gemini may spit out accomplished photographs. Anywhere beneath a minute will get a gold star from me, and Gemini was commonly dealing with requests in beneath 15 seconds. I additionally appreciated the way it added a watermark to all the photographs it created and edited — even when I do not love how tech firms have corrupted the sparkles emoji for AI, it is extraordinarily essential to have some markers of AI-generated content material. Google’s SynthID and behind-the-scenes work additionally assist differentiate AI content material from human-created imagery.Gemini is nice at wholesale AI picture creation, too, however I like to recommend utilizing its Imagen 4 or one other AI picture generator as an alternative — they’ve extra hands-on controls and settings that get you nearer to what you need with much less work.What actually did not workThere are severe limitations to Gemini bananas. It robotically generated sq. photographs, and follow-up prompts asking for photographs to be tailored into different dimensions had been ignored or failed.I additionally seen that Gemini diminished the decision of a lot of my images. I primarily take images with my iPhone 16, which has stellar cameras, however after going by means of the Gemini bananas mannequin, these advantageous particulars had been usually blurred. That’s annoying and will not win over any photographers. Despite my request to brighten up the goal’s rings, the unique picture (left) has extra richness and texture than the AI-edited one (proper). Katelyn Chedraoui/GeminiI attempted repeatedly to get Gemini to deal with picture edits that might’ve been tough for me to do manually. That’s one space in picture modifying the place AI is meant to excel — automating mundane however detail-intensive edits. Sadly, Gemini actually struggled with immediate adherence right here, which means it did not do what I requested. I attempted many instances to get Gemini to take away reflections from a snap of a Freakier Friday film poster, however they stubbornly remained. And the extra I attempted to get it to take away the reflections, the poorer the standard of the picture grew to become with each immediate. Once-clear textual content was in the end illegible after I lastly gave up, to not point out the unintended, scary-looking injury carried out to the faces of Lindsey Lohan and Jamie Lee Curtis.Gemini nano bananas struggled to generate photographs in several dimensions. Resizing and cropping photographs is a core picture modifying course of, however Gemini did not — or could not — deal with easy sizing pointers in my prompts. All I wished was for Gemini to take away the reflection within the upper-right nook of the poster, however as an alternative it warped the film tag line and faces. Katelyn Chedraoui/GeminiI reached out to Google in regards to the decision and dimension points and a spokesperson stated the tech firm is “aware and actively working on both issues. It’s been a big update from our previous model but we’ll continue to improve on the model.”Overall, Gemini nano bananas proved to me that Google is severe about persevering with to dominate in generative media. But it has important pitfalls, with too massive a concentrate on producing new parts relatively than utilizing AI to enhance and tweak widespread picture points. For now, the nano bananas mannequin is finest suited to Gemini followers who wish to make massive edits shortly. For these of us in search of extra exact instruments, we’ll have to attend for Google’s subsequent massive replace or discover one other program.Gemini nano bananas availability, pricing and privatenessYou needn’t do something to entry the brand new mannequin; it is robotically added to the bottom Gemini 2.5 Flash mannequin. Gemini is out there totally free, with extra fashions and better utilization caps out there in Google’s AI plans beginning at $20 per 30 days. If you are a paying subscriber, you might also be capable to entry the mannequin by means of Google AI Studio. From there, all it’s important to do is add a picture and kind out your immediate. Each immediate makes use of wherever from one to 2 thousand tokens, relying on the extent of element required. Adobe Express and Firefly customers may entry the brand new mannequin now. Google’s Gemini privateness coverage says it may well use the data you add for bettering its AI merchandise, which is why the corporate recommends avoiding importing delicate or personal data. The firm’s AI prohibitive use coverage additionally outlaws the creation of unlawful or abusive materials.For extra, take a look at one of the best AI picture mills and the whole lot introduced on the Made by Google Pixel 10 occasion.
More