Google unveils Gemini's "nano banana" model, taking AI image generation to the next level

midian182

midian182

Posts: 11,750   +178
Staff member
What just happened? Google has just unveiled a major upgrade to Gemini AI's image generation capabilities. Gemini 2.5 Flash, a.k.a. "nano banana" has already ranked as the world's top image editor on the LMArena leaderboard and it's earning rave reviews from users. Nano Banana is designed to solve one of AI's biggest frustrations: consistency. By enabling precise edits, multi-turn tweaking, and seamless style mixing, Google isn't just chasing technical polish – it's gunning for its own breakout cultural moment, the kind of mass adoption that Studio Ghibli – style generations once delivered for GPT.

Available in the Gemini app and to developers via the Gemini API, Google AI Studio, and Vertex AI platforms, nano banana is said to address one of the biggest issues with AI image generation: consistency across edits.

If you have an image you like but want to tweak minor details, you've probably experienced the frustration of the entire picture changing when you ask an AI like ChatGPT or Grok to make a small edit.

Google writes that users can, for example, upload a photo of a person and try putting them in different outfits, changing haircuts, or placing them in a different decade, all without distorting the subject in some nightmarish way.

"You can now place the same character into different environments, showcase a single product from multiple angles in new settings, or generate consistent brand assets, all while preserving the subject," the company wrote.

Users can upload a photo of a person and a pet and blend them together in a new scene.

There's also multi-turn editing, which lets you continuously make edits to images. Google suggests adding furniture and decorations to a photo of a room to inspire ideas.

Another interesting element is mixing designs. This lets you apply the style of one image to an object in another, such as changing a dress design to the pattern on a butterfly's wings.

As AI image generators become more advanced and difficult to identify as fake, so do concerns over their use for nefarious purposes. But Gemini 2.5 Flash Image does have the usual AI watermark in the corner, and each image has an invisible SynthID digital watermark that can be detected even if the image has been modified.

Image generation is becoming one of the main battlegrounds in the battle between generative-AI apps. Elon Musk has long championed Grok's abilities in this area, and while most other AIs have guardrails to prevent NSFW images, Grok comes with a "Spicy" mode designed to output this content.

ChatGPT's image-generation abilities helped push its number of users to almost one billion in April, thanks mostly to the massive number of images being created in the style of Studio Ghibli.

Meta, meanwhile, has announced that it is licensing AI image models for Midjourney.

Permalink to story:

Google unveils Gemini's "nano banana" model, taking AI image generation to the next level

 
If it can bring Michael back, this is it.

Jokes aside, I understand the safety concerns, particularly with regards to news, verification, and widespread access. However, in principle, it's no different from film-making, where, before the CGI era, one could not tell the difference in many instances because special effects were done in camera, with models, or with ingenuity. For example, "Alien," or the upside-down fountain in "A Nightmare on Elm Street" after Depp's character gets drawn into the bed. Granted, film material is usually known to be fictional, whereas AI pictures may pass as real, but fundamentally, they are the same: a rendering of reality that was not historical or true.

The problem comes down to "meta" verification, outside the artefact.
 
  • Like
Reactions: Anton Longshot
GeoffreyA said:
If it can bring Michael back, this is it.

Jokes aside, I understand the safety concerns, particularly with regards to news, verification, and widespread access. However, in principle, it's no different from film-making, where, before the CGI era, one could not tell the difference in many instances because special effects were done in camera, with models, or with ingenuity. For example, "Alien," or the upside-down fountain in "A Nightmare on Elm Street" after Depp's character gets drawn into the bed. Granted, film material is usually known to be fictional, whereas AI pictures may pass as real, but fundamentally, they are the same: a rendering of reality that was not historical or true.

The problem comes down to "meta" verification, outside the artefact.
Click to expand...

-Its the democratization of the fakery that's the issue. Want to fake the moon landing? You need tons of set design, the right cameras, wirework etc.

Not something someone in 1969 could really pull off in their garage.

You want to do it today? Just feed an AI some prompts, anyone can do it.

Lies will always travel faster than the truth, and now they don't even have one two or a handful of sources, but 8 billion people who will flood the zone in a way the FSB/KGB/CIA etc never could.

It will be a totally different kind of paradigm shift than social media, which we still haven't actually absorbed and normalized. Like information nukes, AI generated content is going to blast apart the old world media sphere, power structures, and institutions in a way we cannot even begin to predict.

Likely not for the better as well.
 
  • Like
Reactions: Y2JB47, trgz, GeoffreyA and 3 others
GodisanAtheist said:
-Its the democratization of the fakery that's the issue. Want to fake the moon landing? You need tons of set design, the right cameras, wirework etc.

Not something someone in 1969 could really pull off in their garage.

You want to do it today? Just feed an AI some prompts, anyone can do it.

Lies will always travel faster than the truth, and now they don't even have one two or a handful of sources, but 8 billion people who will flood the zone in a way the FSB/KGB/CIA etc never could.

It will be a totally different kind of paradigm shift than social media, which we still haven't actually absorbed and normalized. Like information nukes, AI generated content is going to blast apart the old world media sphere, power structures, and institutions in a way we cannot even begin to predict.

Likely not for the better as well.
Click to expand...
Here's the thing, people have made this argument when Photoshop and other tools came out. You don't need AI to do this and we've had the ability to do this since the first images hit cellophane. This just makes it easier for anyone to do it now, that's all.
 
  • Like
Reactions: trgz and Anton Longshot
With this facility, it won't be long before catfishing (by both men and women) gets exposed to the unsuspecting.
Men and women are not going to know JUST who they are supposed to be dating.
 
Hollywood should start being worried about their being able to retain those UBER high salaries.
Once AI takes over, THEY will be out of work!
 
I was really looking forward to using this facility but then it appears that if you are in the UK, Switzerland or the EU, its blocked because of GDPR. So, I cant even use my own image to create one of these videos. Not even an option to opt out or even confirm my identity. :(
 
You must log in or register to reply here.

Similar threads

Julio Franco
We found 8 Nano Banana Pro images that show Google is the new benchmark in AI art
Replies
10
Views
154
Liranan
L
S
Google upgrades Gemini image generation to Nano Banana 2 with 4K support
Replies
2
Views
190
Xenocea
Xenocea
midian182
Google Gemini's nano banana AI image generator is blowing up online - take a look at what it can do
2
Replies
27
Views
918
toooooot
T

Latest posts

Back