Anyone know how it handles '1920s nazi officer'? They stopped doing humans for a while but now I see they're back so I wonder how they're handling the criticism they got from that
That lamp example is pretty impressive (though it's hard to know how cherry-picked it is). The lamp is plugged in, it's lighting the things in the scene, it's casting shadows.
I have a certain use case for such image generators. Feed them an entire news article I fetch from bbc and ask it to create an image to accompany the article. Thus far only midjourney managed to understand context. And now this, which is even more impressive. We live in interesting times.
Seems to be failing at API Calls right now with "You exceeded your current quota, please check your plan and billing details. For more information on this error,"
I love that it's substantially faster than ChatGPT's image generation. It takes ages, so slow that the app tells you to not wait and sends you notification when the generation finishes.
I have to say while I'm deeply impressed by these text to image models, there's a part of me that's also wary of their impact. Just look at the comments beneath the average Facebook post.
I think it's time to build a new system - something that can annotate the post the user is on, if there's at least another savvy user (or AI system) that can pick up on the uncanny signals. This youtube video about the "Walker Family" sham on Facebook is particularly relevant here:
I've had a task in mind for a while now that I've wanted to do with this latest crop of very capable instruction-following image editors.
Without going into detail, basically the task boils down to, "generate exactly image 1, but replace object A with the object depicted in image 2."
Where image 2 is some front-facing generic version, ideally I want the model to place this object perfectly in the scene, replacing the existing object, that I have identified ideally exactly by being able to specify its position, but otherwise by just being able to describe very well what to do.
For models that can't accept multiple images, I've tried a variation where I put a blue box around the object that I want to replace, and paste the object that I want it to put there at the bottom of the image on its own.
I've tried some older models, and ChatGPT, also qwen-image last week, and just now, this one. They all fail at it. To be fair, this model got pretty damn close, it replaced the wrong object in the scene, but it was close to the right position, and the object was perfectly oriented and lit. But it was wrong. (Using the bounding box method.. it should have been able to identify exactly what I wanted to do. Instead it removed the bounding box and replaced a different object in a different but close-by position.)
Are there any models that have been specifically trained to be able to infill or replace specific locations in an image with reference to an example image? Or is this just like a really esoteric task?
So far all the in-filling models I've found are only based on text inputs.
Not sure what your exact task is, but I have a similar goal as well. Haven't had time to try alot of different models or ideas yet because got busy with other stuff, but have you tried this: https://youtu.be/dQ-4LASopoM?si=e33FQd5f4fYr4J5L&t=299
where you stitch two images together, one is the working image (the one you want to modify), and the other one is the reference image, you then instruct the model what to do. I'm guessing this approach is as brittle as the other attempts you've tried so far, but I thought this seemed like an interesting approach.
An image seems to be 256 tokens looking the AIstudio tab, so you can generate 3906,25 images per 1M tokens, that seems a lot if I'm not wrong in some ways.
Edit: the blog post is now loading and reports "1290 output tokens per image" even though on the AI studio it said something different.
Another nitpick - the pink puffer jacket that got edited into the picture is not the same as the one in the reference image - it's very similar but if I were to use this model for product placement, or cared about these sort of details, I'd definitely have issues with this.
Regardless, it seems Google is on the frontier of every type of model and robotics (cars). It’s nutty how we forget what a intellectual juggernaut they are.
Before AI, people complained that Google was taking world class engineering talent and using it for little more than selling people ads.
But look at that example. With this new frontier of AI, that world class engineering talent can finally be put to use…for product placement. We’ve come so far.
“Nano banana” is probably good, given its score on the leaderboard, but the examples you show don't seem particularly impressive, it looks like what Flux Kontext or Qwen Image do well already.
I digitised our family photos but a lot of them were damaged (shifted colours, spills, fingerprints on film, spots) that are difficult to correct for so many images. I've been waiting for image gen to catch up enough to be able to repair them all in bulk without changing details, especially faces. This looks very good at restoring images without altering details or adding them where they are missing, so it might finally be time.
This model is very impressive. Yesterday (as nano-banana) I gave it a photo of an indoor scene with a picture hanging on a wall, and asked it the picture on a wall with a copy of the whole photo. It worked perfectly the first time.
It didn't succeed in doing the same recursively, but it's still clearly a huge advance in image models.
125 comments
[ 3.5 ms ] story [ 118 ms ] threadhttps://developers.googleblog.com/en/introducing-gemini-2-5-...
It seems like this is 'nano-banana' all along
This is why I'm sticking mostly to Adobe Photoshop's AI editing because there are no restrictions in that regard.
Hope they get API issues resolved soon.
I have to say while I'm deeply impressed by these text to image models, there's a part of me that's also wary of their impact. Just look at the comments beneath the average Facebook post.
Don’t Pay This AI Family To Write You a Song - https://www.youtube.com/watch?v=u-DDHSfBBeo
Without going into detail, basically the task boils down to, "generate exactly image 1, but replace object A with the object depicted in image 2."
Where image 2 is some front-facing generic version, ideally I want the model to place this object perfectly in the scene, replacing the existing object, that I have identified ideally exactly by being able to specify its position, but otherwise by just being able to describe very well what to do.
For models that can't accept multiple images, I've tried a variation where I put a blue box around the object that I want to replace, and paste the object that I want it to put there at the bottom of the image on its own.
I've tried some older models, and ChatGPT, also qwen-image last week, and just now, this one. They all fail at it. To be fair, this model got pretty damn close, it replaced the wrong object in the scene, but it was close to the right position, and the object was perfectly oriented and lit. But it was wrong. (Using the bounding box method.. it should have been able to identify exactly what I wanted to do. Instead it removed the bounding box and replaced a different object in a different but close-by position.)
Are there any models that have been specifically trained to be able to infill or replace specific locations in an image with reference to an example image? Or is this just like a really esoteric task?
So far all the in-filling models I've found are only based on text inputs.
where you stitch two images together, one is the working image (the one you want to modify), and the other one is the reference image, you then instruct the model what to do. I'm guessing this approach is as brittle as the other attempts you've tried so far, but I thought this seemed like an interesting approach.
Edit: the blog post is now loading and reports "1290 output tokens per image" even though on the AI studio it said something different.
Sorry, there seems to be an error. Please try again soon.”
Never thought I would ever see this on a google owned websites!
Just search nano banana on Twitter to see the crazy results. An example. https://x.com/D_studioproject/status/1958019251178267111
But look at that example. With this new frontier of AI, that world class engineering talent can finally be put to use…for product placement. We’ve come so far.
“Nano banana” is probably good, given its score on the leaderboard, but the examples you show don't seem particularly impressive, it looks like what Flux Kontext or Qwen Image do well already.
It didn't succeed in doing the same recursively, but it's still clearly a huge advance in image models.