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

4o Image Generation

openai.com

Post reply on HN
Print view View original

381–390 of 629 posts

Re: 4o Image Generation

#381

Post by low_tech_love »

Taek wrote: post #242
minimaxir wrote: post #2

OpenAI's livestream of GPT-4o Image Generation shows that it is slowwwwwwwwww (maybe 30 seconds per image, which Sam Altman had to spin "it's slow but the generated images are worth it"). Instead of using a diffusion approach, it appears to be generating the image tokens and decoding them akin to the original DALL-E ( https://openai.com/index/dall-e/ ), which allows for streaming partial generations from top to botto…

> it appears to be generating the image tokens and decoding them akin to the original DALL-E The animation is a lie. The new 4o with "native" image generating capabilities is a multi-modal model that is connected to a diffusion model. It's not generating images one token at a time, it's calling out to a multi-stage diffusion model that has upscalers. You can ask 4o about this yourself, it seems to have a strong under…

Would it seem otherwise if it was a lie?

Re: 4o Image Generation

#382

Post by nestorD »

jay_kyburz wrote: post #322
nmilo wrote: post #288

Visual internet content is completely over. Pack it up

So I spent a good few hours investigating the current state of the art a few weeks ago. I would like to generate a collection of images for the art in a video game. It is incredibly difficult to develop an art style, then get the model to generate a collection of different images in that unique art style. I couldn't work out how to do it. I also couldn't work out how to illustrate the same characters or objects in di…

Short answer: the model is good at consistency. You can use it to generate a set a style reference images, then use those as reference for all your subsequent generations. Generating in the same chat might also help it have further consistency between images.

Re: 4o Image Generation

#383

Post by low_tech_love »

mgraczyk wrote: post #380
low_tech_love wrote: post #378

Earlier quoted context omitted.

Honest question, do you believe something just because the bot tells you that?

No, did you look at my link?

Yes, and it shows you believing what the bot is telling you, therefore I asked. It is giving you some generic function call with a generic name. Why would you believe that is actually what happens with it internally?

By the way when I repeated your prompt it gave me another name for the module.

Re: 4o Image Generation

#384

Post by nestorD »

wiradikusuma wrote: post #379

Just curious if it works for creating a comic strip? I.e. will it maintain the consistency of the characters? I watched a video somewhere they demo'ed it creating comic panels, but I want to create the panels one by one.

I believe so! Since it is good at consistency and can be feed reference images, you can generate character references and deed those, along with the previous panels, to the model working one panel at a time.

Re: 4o Image Generation

#385

Post by mgraczyk »

low_tech_love wrote: post #383
mgraczyk wrote: post #380

Earlier quoted context omitted.

No, did you look at my link?

Yes, and it shows you believing what the bot is telling you, therefore I asked. It is giving you some generic function call with a generic name. Why would you believe that is actually what happens with it internally? By the way when I repeated your prompt it gave me another name for the module.

Please share your chat

I also just confirmed via the API that it's making an out of band tool call

EDIT: And googling the tool name I see it's already been widely discussed on twitter and elsewhere

Re: 4o Image Generation

#386

Post by DeathArrow »

blixt wrote: post #104

What's important about this new type of image generation that's happening with tokens rather than with diffusion, is that this is effectively reasoning in pixel space. Example: Ask it to draw a notepad with an empty tic-tac-toe, then tell it to make the first move, then you make a move, and so on. You can also do very impressive information-conserving translations, such as changing the drawing style, but also stuff l…

>You can also do very impressive information-conserving translations, such as changing the drawing style, but also stuff like "change day to night", or "put a hat on him", and so forth.

You can do that with diffusion, too. Just lock the parameters in ComfyUi.

Re: 4o Image Generation

#387

Post by hansvm »

cubefox wrote: post #49
minimaxir wrote: post #2

OpenAI's livestream of GPT-4o Image Generation shows that it is slowwwwwwwwww (maybe 30 seconds per image, which Sam Altman had to spin "it's slow but the generated images are worth it"). Instead of using a diffusion approach, it appears to be generating the image tokens and decoding them akin to the original DALL-E ( https://openai.com/index/dall-e/ ), which allows for streaming partial generations from top to botto…

LLMs are autoregressive, so they can't be (multi-modality) integrated with diffusion image models, only with autoregressive image models (which generate an image via image tokens). Historically those had lower image fidelity than diffusion models. OpenAI now seems to have solved this problem somehow. More than that, they appear far ahead of any available diffusion model, including Midjourney and Imagen 3. Gemini "int…

> so they can't be integrated

That's overly pessimistic. Diffusion models take an input and produce an output. It's perfectly possible to auto-regressively analyze everything up to the image, use that context to produce a diffusion image, and incorporate the image into subsequent auto-regressive shenanigans. You'll preserve all the conditional probability factorizations the LLM needs while dropping a diffusion model in the middle.

Re: 4o Image Generation

#388

Post by eitland »

meeton wrote: post #185
jjbinx007 wrote: post #130

Earlier quoted context omitted.

It still can't generate a full glass of wine. Even in follow up questions it failed to manipulate the image correctly.

https://i.imgur.com/xsFKqsI.png "Draw a picture of a full glass of wine, ie a wine glass which is full to the brim with red wine and almost at the point of spilling over... Zoom out to show the full wine glass, and add a caption to the top which says "HELL YEAH". Keep the wine level of the glass exactly the same."

Most interesting thing to me is the spelling is correct.

I'm not a heavy user of AI or image generation in general, so is this also part of the new release or has this been fixed silently since last I tried?

Re: 4o Image Generation

#389

Post by gloosx »

ASalazarMX wrote: post #352
dghlsakjg wrote: post #326

Earlier quoted context omitted.

The head of foam on that glass of wine is perfect !

I think we're really fscked, because even AI image detectors think the images are genuine. They look great in Photoshop forensics too. I hope the arms race between generators and detectors doesn't stop here.

We're not. This PNG image of a wine glass has JPEG compression artefacts which are leaking from JPEG training data. You can zoom into the image and you will see 8x8 boundaries of the blocks used in JPEG compression, which just cannot be in a PNG. This is a common method to detect AI-generated image and it is working so far, no need for complex photoshop forensics or AI-detectors, just zoom-in and check for compression - current AI is incapable of getting it right – all the compression algorithms are mixed and mashed in the training data, so on the generated image you can find artefacts from almost all of them if you're lucky, but JPEG is prevalent obviously, lossless images are rare online.

Re: 4o Image Generation

#390

Post by jychang »

outworlder wrote: post #291
blixt wrote: post #268

Earlier quoted context omitted.

This argument could be made for every level of abstraction we've added to software so far... yet here we are commenting about it from our buggy apps!

Yeah, but the abstractions have been useful so far. The main advantage of our current buggy apps is that if it is buggy today, it will be exactly as buggy tomorrow. Conversely, if it is not currently buggy, it will behave the same way tomorrow. I don't want an app that either works or does not work depending on the RNG seed, prompt and even data that's fed to it. That's even ignoring all the absurd computing power th…

You're just describing calling a customer service phone line in India.
Post reply on HN