I’ve been trying Google’s latest image editing features Generated AI Model, so can you. Named Gemini 2.5 Flash Images, IT upgrade The ability to edit photos locally in Gemini.
AI enthusiasts call it the “nanobanana” model, stimulated by a series of trailers Google implement. Google has been investing heavily in its generative media models this year, removing updated versions on its image and video generator models Annual I/O Developer Conference. Google’s AI Video Generator VEO 3 Shocked by synchronizing audio, this is the first of the AI giants. Creators are more than just 100 million AI videos Use Google’s AI filmmaker tool Flow.
I spent a lot of time testing AI creative software and was happy to see Google’s latest in all the banana carnivals. But my tests of 2.5 Flash images show that just because of something’s entrance glitz doesn’t mean it always fits its hype. Here’s my experience with Gemini nanobananas: good, bad and frustrating.
What works
The Gemini Bananas model is very good at adding elements to existing images, blending AI-generated elements well into any image you take. It also maintains a steady level of character consistency – which means the people in my photos aren’t too distorted or weird after being AI-treated. These are all important differences in AI image programs, and Google says it has helped improve.
You can see these two features in this photo of my sister and me. Our general appearance has not changed in the edited version (right), showing the consistency of the character. I asked Gemini to add a sister who looks similar to our two, which was excellent by adding a third woman between the two of us.
Welcome to the family, it looks very precise.
I was also impressed by the speed at which Gemini was able to spit out a complete image. In less than a minute, I will get Venus from me, and Gemini regularly process requests within 15 seconds. I also appreciate how it adds watermarks to all the images it creates and edits – even if I don’t like tech companies Spelling ai’s spark emojihave some very important Tags of content generated by AI. Google’s synthesis and behind-the-scenes work also helps distinguish AI content from artificially created images.
Gemini is also good at wholesale AI image creation, but I recommend using its Imaging 4 or other images AI image generator Instead, they have more hands-on controls and settings that bring you closer to the job you want with less work.
What really doesn’t work
Gemini bananas have serious limitations. It automatically generates a square image and ignores or fails, requiring the image to be adapted to other dimensions.
I also noticed that Gemini reduced the resolution of many photos. I mainly take pictures with my iPhone 16, which has a great camera, but these details usually get blurred after passing the Gemini Banana model. It’s annoying and won’t win any photographers.
Although I asked for a ring to improve my goal, the original photo (left) has a richer richer texture than the AI-edited photo (right).
I’ve tried repeatedly to have Gemini handle photo editing, which is hard for me to do manually. This is one area in photo editing that AI should perform well – automating mundane but detailed-intensive editing. Sadly, Gemini does stick quickly here, which means it doesn’t do what I asked.
I’ve tried many times to get Gemini to remove reflections from snapshots of a weird Friday movie poster, but they still stubbornly exist. And, the more I try to remove the reflection, the worse the quality of the image, and each prompt will be more and more. After I finally gave up, the once clear text was ultimately unrecognizable, not to mention by chance, the horrible damage caused Lindsey Lohan and Jamie Lee Lee Curtis’ faces.
Gemini Nano Bananas strives to produce images on different dimensions. Resizing and cropping images is a core photo editing process, but Gemini doesn’t (or can’t) handle simple sizing guides in my tips.
All I wanted was Gemini to remove the reflection in the upper right corner of the poster, but instead distorting the movie tag line and face.
I contacted Google for information about solutions and dimension issues, and a spokesperson said the tech company “aware of and actively researched both issues. This is a major update to our previous model, but we will continue to improve the model.”
Overall, Gemini Nano Bananas proved to me that Google takes it seriously and continues to dominate in generative media. But it has a big pitfall, with the focus on generating new elements rather than using AI to improve and adjust common photo issues. Currently, the Nano Bananas model is best for Gemini fans who want to do a large-scale edit quickly. For those looking for more precise tools, we have to wait for Google’s next big update or find another program.
Gemini Nanobanana Availability, Pricing and Privacy
You don’t need to do anything to access the new model; it will be automatically added to the Base Gemini 2.5 Flash model. Gemini is available for free, and more models and higher usage caps are available in the Google AI plan. $20 per month.
If you are a paid subscriber, you can also access the model through Google AI Studio. From there, all you have to do is upload the image and enter the prompt. Each prompt uses one to two thousand tokens, depending on the level of detail required. Adobe Express and Firefly users also have immediate access to the new model.
Google’s Gemini Privacy Policy It means that the information you upload can be used to improve its AI products, which is why companies recommend avoiding uploading sensitive or private information. Company’s Artificial Intelligence Usage Policy The creation of illegal or abused materials has also been cancelled.
For more information, please check it out Best AI image generator and All about Google Pixel 10 Events.
