fwiw i've added a summary of this discussion here (https://extraakt.com/extraakts/gpt-4o-image-generation-discu...) to keep track of the main points
4o Image Generation
561–570 of 629 posts
Re: 4o Image Generation
#562Earlier quoted context omitted.
>First AI image generator to pass the uncanny valley test? Other image generators I've used lately often produced pretty good images of humans, as well [0]. It was DALLE that consistently generated incredibly awful images. Glad they're finally fixing it. I think what most AI image generators lack the most is good instruction following. [0] YandexArt for the first prompt from the post: https://imgur.com/a/VvNbL7d The…
Do you have another example from YandexArt? https://images.ctfassets.net/kftzwdyauwt9/7M8kf5SPYHBW2X9N46... OpenAI's human faces look *almost* real.
>OpenAI's human faces look almost real.
Not sure, I tried a few generations, and it still produces those weird deformed faces, just like the previous generation: https://imgur.com/a/iKGboDH Yeah, sometimes it looks okay.
YandexArt for comparison: https://imgur.com/a/K13QJgU
Re: 4o Image Generation
#563Earlier quoted context omitted.
No, I wasn't planning to post this for a couple weeks, but I saw the comment and was like "eh, why not?". You can watch "sped up" past sessions by other people who used this demo here, which is kind of like a demo video: https://universal.oroborus.org/gallery But the gallery feature isn't really there today, it shows all the "one-click and bounce sessions", and its hard to find signal in the noise. I'll probably subm…
Seconding the need for a video. We need a way to preview this without it costing you money. I had to charge you a few dimes to grasp this excellent work. The description does not do it justice; people need to see this in motion. The progressive build-up of a single frame, too. I encourage you to post the Show HN soon.
Anyone know order-or-magnitude how many visits to expect (order of magnitude) from an Ask HN? A thousand? 10? 100? I need to figure out how many credits I'd need to line up to survive one.
> had to charge you a few dimes
s/you/openrouter/: ty to openrouter for donating a significant chunk of credits a couple hours ago.
Really appreciate the feedback on needing a video. I had a sense this was the most important "missing piece", but this will give me the motivation to accomplish what is (to me) a relatively boring task, compared to hacking out more features.
Re: 4o Image Generation
#564Earlier quoted context omitted.
I mean I literally "tried it again" this morning, as a paying Copilot customer of 12 months, to the result I already described. And I do not want to "try it" - based on fluffy promises we've been hearing, it should "just work". Are you old enough to remember that phrase? It was a motto introduced by an engineering legend whose devices you're likely using every day. The reason why "everyone", including myself with 20+…
Steve Jobs may be a legend at business, but an engineer he is not. To say nothing of the fact that whole reason "it just works" is because of said engineering. If you would like to be the innovator that finally solves that, then great! Otherwise you're just bloviating, and by god do we already have enough of that in this field. I'm approaching 20 years of professional SWE experience myself. The boring shit is my brea…
The point is not the individual tools, which at this point are just wrappers around the major LLMs. The point is the snake oil salesmen of major LLM companies have been telling us for several years now that it is "just about to happen". A new technology revolution. A new post-scarcity world if you will. A tremendous increase in technological output, unleashed creativity etc. Altman routinely blabs about achieving AGI. Meanwhile hallucination of the models is a known feature, unfortunately it's not a bug we can fix. The hallucinations will never go away, because the LLM models are advanced text generators (quoting Charles Petzold) producing text based on essentially probability that one token should follow the next one. That means mate, you can be a superstar at the "advanced" skill of "prompting" (i.e. typing in conversational english language sentences), the crappy tool will still produce output that does not make sense, for example type out code with non-existing framework methods etc. Why? Because with every prompt, you retrain and re-tune the model a little bit. They don't even hold the authority of a dusty old encyclopedia. You use several tools simultaneously, why? Because you cannot really rely on any of them. So you try to get a mean minimum out of several. But a mean minimum of a sum of crap, will still end up crap. If any of the the 3-4 major LLM engines had any competitive advantages, they would have literally obliterated the competition by now. Why is that not happening? Where is the LLM equivalent of nascent Google obliterating Altavista and Excite or an equivalent of Windows 95 taking the PC, React taking over the web frontend etc? And by the way, you know that there was another famous guy at Apple, right?
Re: 4o Image Generation
#565Earlier quoted context omitted.
No, I wasn't planning to post this for a couple weeks, but I saw the comment and was like "eh, why not?". You can watch "sped up" past sessions by other people who used this demo here, which is kind of like a demo video: https://universal.oroborus.org/gallery But the gallery feature isn't really there today, it shows all the "one-click and bounce sessions", and its hard to find signal in the noise. I'll probably subm…
Seconding the need for a video. We need a way to preview this without it costing you money. I had to charge you a few dimes to grasp this excellent work. The description does not do it justice; people need to see this in motion. The progressive build-up of a single frame, too. I encourage you to post the Show HN soon.
The main things I'm blocking on before submitting a "Show HN" are:
- Getting the instant-replay gallery sorted to make it usable, maybe sorting via likes.
- Selecting a couple interesting sessions from the gallery and turning them into a short video
- Making sure I have enough credits lined up (hopefully donations!) to survive an "Ask HN".
Re: 4o Image Generation
#566Earlier quoted context omitted.
I mean I literally "tried it again" this morning, as a paying Copilot customer of 12 months, to the result I already described. And I do not want to "try it" - based on fluffy promises we've been hearing, it should "just work". Are you old enough to remember that phrase? It was a motto introduced by an engineering legend whose devices you're likely using every day. The reason why "everyone", including myself with 20+…
None of this stuff even existed 3 years ago and you're asking like we're talking about self-driving cars. What hubris. My god.
There is no hubris. The LLM technology has actually been existing for at least two decades, it's not some sudden breakthrough we suddenly discovered. And given the many billions of dollars it has sucked in, it's definitely a pile of crap. I have been a paying customer of Github Copilot for at least a year. Since the google search has been completely messed up, sometimes it can be useful to look up some cryptic error message. It can also sometimes help recall syntax of something. But it's not the magic machine they've been touting, it's definitely not the AGI, and my god, it's very prone to errors and to anyone who admires this tech, please for the love of god double- and tripple-check the crap that they generate, before you commit it to production. And by now it definitely feels like the 'self-driving' cars 'revolution'. They have been 'just around the corner', for like what, 15 years now?
Re: 4o Image Generation
#567Ran through some of my relatively complex prompts combined with using pure text prompts as the de-facto means of making adjustments to the images (in contrast to using something like img2img / inpainting / etc.) https://mordenstar.com/blog/chatgpt-4o-images It's definitely impressive though once again fell flat on the ability to render a 9-pointed star.
Have u had any luck with engineering/schematics/wireframe diagrams such as [1] ?? [1] https://techcrunch.com/wp-content/uploads/2024/03/pasted-ima...
Great question. I haven't tested the creation of such an image from scratch, but I did add an adjustment test against that specific text-heavy diagram and I'd say it passed with "flying colors". (pun intended).
Re: 4o Image Generation
#568Earlier quoted context omitted.
None of this stuff even existed 3 years ago and you're asking like we're talking about self-driving cars. What hubris. My god.
There is no hubris. The LLM technology has actually been existing for at least two decades, it's not some sudden breakthrough we suddenly discovered. And given the many billions of dollars it has sucked in, it's definitely a pile of crap. I have been a paying customer of Github Copilot for at least a year. Since the google search has been completely messed up, sometimes it can be useful to look up some cryptic error…
I’m sorry but you will be the first to go in this new age. LLMs today are absolutely mind blowing, along with this image stuff. Either learn to adapt or remain a boomer.
Re: 4o Image Generation
#569Earlier quoted context omitted.
None of this stuff even existed 3 years ago and you're asking like we're talking about self-driving cars. What hubris. My god.
There is no hubris. The LLM technology has actually been existing for at least two decades, it's not some sudden breakthrough we suddenly discovered. And given the many billions of dollars it has sucked in, it's definitely a pile of crap. I have been a paying customer of Github Copilot for at least a year. Since the google search has been completely messed up, sometimes it can be useful to look up some cryptic error…
No, LLMs have not exited for two decades. What a stupid comment. Millions of people are spending thousands of dollars because they're recieving tens of thousands of dollars of value from it.
Of course it's impossible to explain that to thickheaded dinosaurs on HN who think they're better than everyone and god's gift to IQ.
Re: 4o Image Generation
#570Earlier quoted context omitted.
There is no hubris. The LLM technology has actually been existing for at least two decades, it's not some sudden breakthrough we suddenly discovered. And given the many billions of dollars it has sucked in, it's definitely a pile of crap. I have been a paying customer of Github Copilot for at least a year. Since the google search has been completely messed up, sometimes it can be useful to look up some cryptic error…
I’m sorry but you will be the first to go in this new age. LLMs today are absolutely mind blowing, along with this image stuff. Either learn to adapt or remain a boomer.
Go where mate, to Sam Altman's retirement home for UBI recipients ? I studied neural networks while you were still a "concept of a plan" in the mind of your parents and unlike you, I know how they work. As one of the early and paying adopters of the technology, it would have been great if they worked as advertised. But they don't, and the only people "to go" will be idiots who think that a technology that 1) anyone can use 2) produces un-reliable outputs.. 3)..while sounding authoritative makes them an expert. Guess what, if both you and your buddy and your entire school can spin up a website with a few prompts, how much is your "skill" worth on the market? Ever heard of demand and offering? To me looks eerily similar to how the smartphones and social networks made everyone "technologists" ;)