Live data from Hacker News

GPT Image 1.5

openai.com

241–250 of 272 posts

Re: GPT Image 1.5

#241

Earlier quoted context omitted.

There are no watermarks in the arena.

There are no visible watermarks, but model makers can use steganographic codes to identify outputs from their own models.

Text-to-Image Models Leave Identifiable Signatures: Implications for Leaderboard Security

https://arxiv.org/pdf/2510.06525

Re: GPT Image 1.5

#242

Okay results are in for GenAI Showdown with the new gpt-image 1.5 model for the editing portions of the site! https://genai-showdown.specr.net/image-editing Conclusions - OpenAI has always had some of the strongest prompt understanding alongside the weakest image fidelity. This update goes some way towards addressing this weakness. - It's leagues better at making localized edits without altering the entire image's ae…

I disagree with gpt-image-1.5's grade on the worm sign. It moved some of the marks around to accommodate the enlarged black area, but retained the overall appearance of the sign.

I can see how you'd come to that conclusion. Each prompt is supposed to illustrate a different type of test criteria. The ultimate goal of Worm Sign is intended to test a near 100% retention of the original weathered/dented sign.

If you look at the ones that passed (Flux.2 Pro, Gemini 2.5 Flash, Reve), you'll see that they did not add/subtract/move any of the pockmarks from the original image.

Re: GPT Image 1.5

#243

Okay results are in for GenAI Showdown with the new gpt-image 1.5 model for the editing portions of the site! https://genai-showdown.specr.net/image-editing Conclusions - OpenAI has always had some of the strongest prompt understanding alongside the weakest image fidelity. This update goes some way towards addressing this weakness. - It's leagues better at making localized edits without altering the entire image's ae…

Absolutely fabulous work. Ludicrously unnecessary nitpick for "Remove all the brown pieces of candy from the glass bowl": > Gemini 2.5 Flash - 18 attempts - No matter what we tried, Gemini 2.5 Flash always seemed to just generate an entirely new assortment of candies rather than just removing the brown ones. The way I read the prompt, it demands that the candies should change arrangement. You didn't say "change the b…

That is a great point! Since we are moving towards better "world models" in terms of these multimodal models, you could reasonably argue that if the directive was to physically remove the candy that in the process of doing so, gravity/physics could affect the positioning of other objects.

You will note that the Minimum Passing Criteria allows for a color change in order to pass the prompt but with the rapid improvements in generative models, I may revise this test to be stricter, only allowing "Removal" to be considered as pass as opposed to a simple color swap.

Re: GPT Image 1.5

#244

Earlier quoted context omitted.

If somebody is stealing from your bank account every week and you just don’t notice it, are you not being stolen from? Has nobody stolen your credit card and used it until the moment you notice the charges. I don’t really think we can go “if a tree fall in the forest and nobody is around to hear it…” about this. Stallman has his opinions on software, I have my opinions on my visual work. I don’t get really how that a…

If someone steals from my bank account I certainly CAN notice it even if I don't immediately, and I'm certainly worse off. That's such a bad straw man I wonder if you're really supporting the position you claim to be supporting. Maybe you're just trying to give it a bad name. Your opinion isn't on visual work, but visual property. You don't demand to be paid for your work - your labor. Rather you traded that for the…

If you think that’s a bad example so be it but I’m not attempting to make a strawman or give anything a bad name.

I don’t really know where all the hostility came from in this conversation but I think it’s best if we move on.

Re: GPT Image 1.5

#245

Earlier quoted context omitted.

As a professional cinematographer/photographer I am incredibly uncomfortable with people using my art without my permission for unknown ends. Doubly so when it’s venture backed private companies stealing from millions of people like me as they make vague promises about the capabilities of their software trained on my work. It doesn’t take much to understand why that makes me uncomfortable and why I feel I am entitled…

You should be proud your work will now be distilled enterally and an aspect of your work will forever influence the world

I’m not

Re: GPT Image 1.5

#246

I still use Midjourney, because all of these major players are so bad at stylistic and creative work. They're singularly focused on photorealism.

In my experience, MidJourney creates the best overall-looking images, but it's the worst at sticking to your prompt.

Re: GPT Image 1.5

#247

Earlier quoted context omitted.

They're still fine because they're right. You got to play the copyright game when the big corps were on your side. Now they're on the other side. Deal with it and get over it.

You are not entitled to my art. Comparing that to copyright abuse by large corporations is ridiculous.

I get access to inspiration from everybody's art, and so do you. Seems like a good deal to me.

Meanwhile, the next generation of great artists is already at work down the street from you. Some kids you've never heard of, playing around in a basement or garage you've probably driven past a hundred times. They're learning to make the most of the tools at hand, just like the old masters did. Except the tools at hand this time are little short of godlike.

It's an exciting time. If you wanted things to stay the same, you shouldn't have gone into technology or art.

Re: GPT Image 1.5

#248
post #196

Earlier quoted context omitted.

My pet theory is that OpenAI screwed up the image normalization calculation and was stuck with the mistake since that's something that can't be worked around. At the least, it's not present in these new images.

wdym it cant be worked around when there exist literal yellow tint corrector models/tools haha

There's a possibility that any automatic correction could have false positives (since the yellow tint doesn't happen 100% of the time) which creates different problems where a image could have an even weirder hue.

Re: GPT Image 1.5

#249

Earlier quoted context omitted.

AI doesn’t have much of a moat. People can and will easily switch providers.

Sure but there are only a couple leading providers worth considering for coding at least, and there will be consolidation once investment pulls back. They may find a way to collude on raising prices. Where switching will be easier is with casual chat users plus API consumers that are already using substandard models for cost efficiency. But there will also always be a market for state of art quality.

Reinforced today:

As Gemini has gained competitiveness (higher confidence in its output, better reputation), its prices have steadily risen

Re: GPT Image 1.5

#250
post #240

Earlier quoted context omitted.

What you want, and what you think image generation is, is impossible.

And yet we can see Gemini do what I wanted, so it's clearly not impossible.

What you've found is a prompt that returns what you want on Gemeni. That's all.
Post reply on HN