Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

GPT Image 1.5

openai.com

Post reply on HN
Print view View original

251–260 of 272 posts

Re: GPT Image 1.5

#251

Post by yuni_aigc »

One thing I’ve noticed when comparing these models is that “quality” and “realism” don’t always move together.

Some models are very strong at sharp details and localized edits, but they can break global lighting consistency — shadows, reflections, or overall scene illumination drift in subtle ways. GPT-Image seems to trade a bit of micro-detail for better global coherence, especially in lighting, which makes composites feel more believable even if they’re not pixel-perfect.

It’s hard to capture this in benchmarks, but for real-world editing workflows it ends up mattering more than I initially expected.

Re: GPT Image 1.5

#252

Post by nicpottier »

vunderba wrote: post #102

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…

Love this benchmark, always the first place I look. Also seems like it is time to move the goalposts, not sure we are getting enough resolution between models anymore.

Out of curiosity why does gemini get gold for the poker example but gpt-image 1.5 does not? I couldn't see a difference between the two.

Re: GPT Image 1.5

#253

Post by encroach »

nycdatasci wrote: post #241
nycdatasci wrote: post #222

Earlier quoted context omitted.

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

This is true, however LMArena does employ some methods to mitigate attempts to manipulate the leaderboard, see https://openreview.net/forum?id=zf9zwCRKyP

They also control for style https://news.lmarena.ai/sentiment-control/

Re: GPT Image 1.5

#254

Post by eterm »

bongodongobob wrote: post #250
eterm wrote: post #240

Earlier quoted context omitted.

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.

It's a prompt I've been using for years. Gemini has been the best of the bunch, but Nana Banana, midjourney, etc, all did okay to various degrees.

GPT Image bombed notably worse than the others, not the original picture itself, but the complete lack of recognition of my feedback that it hadn't got it right, it just doubled down on the image it had generated.

Re: GPT Image 1.5

#256

Post by panopticon »

realharo wrote: post #225
rafram wrote: post #108

Earlier quoted context omitted.

That seems unlikely to me. One side is made up of lots and lots of entrenched interests with sympathetic figures like authors and artists on their side, and the other is “big tech,” dominated by the rather unsympathetic OpenAI and Google.

The other side however has the "if you restrict us, China will win" argument on their side.

That argument is easy to politicize and selectively ignore. See: renewables and EVs.

Re: GPT Image 1.5

#257

Post by Bombthecat »

vunderba wrote: post #102

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 can't click the compliance info button on mobile. The text shows for half a second and then vanishes. Long press just marks the text for copy paste.

Re: GPT Image 1.5

#258

Post by Forgeties79 »

CamperBob2 wrote: post #247
Forgeties79 wrote: post #234

Earlier quoted context omitted.

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 thi…

Inspiring artists =/= involuntarily training privately owned LLM’s that charge for access.

If you want me to hand some of my work over to artists so they can learn and grow and experiment, send them my way. Happy to help.

Re: GPT Image 1.5

#259

Post by whywhywhywhy »

nobody_r_knows wrote: post #53
oxag3n wrote: post #46

If this was a farm of sweatshop Photoshopers in 2010, who download all images from the internet and provide a service of combining them on your request, this would escalate pretty quickly. Question: with copyright and authorship dead wrt AI, how do I make (at least) new content protected? Anecdotal: I had a hobby of doing photos in quite rare style and lived in a place where you'd get quite a few pictures of. When I…

my question to your anecdotal: who cares? not being fecicious, but who cares if someone reproduced your stuff and millions of people see your stuff? is the money that you want? is it the fame? because fame you will get, maybe not money... but couldn't there be another way?

The people building the tech are extremely fussy about their work being cited and extremely protective of their models files so they themselves have massive issues with their work being used or replicated non-consensually.

Re: GPT Image 1.5

#260

Post by CamperBob2 »

Forgeties79 wrote: post #258
CamperBob2 wrote: post #247

Earlier quoted context omitted.

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 thi…

Inspiring artists =/= involuntarily training privately owned LLM’s that charge for access. If you want me to hand some of my work over to artists so they can learn and grow and experiment, send them my way. Happy to help.

Inspiring artists =/= involuntarily training privately owned LLM’s that charge for access.

Agreed there, which is why it's important to work for open access to the results. The resulting regime won't look much like present-day copyright law, but if we do it right, it will be better for us all.

In other words, instead of insisting that "No one can have this," or "Only a few can have this," which (again) will not be options for works that you release commercially, it's better IMHO to insist that "Everyone can have this."

Post reply on HN