Hacker News Reader

Comment by hospitalJail | Hacker News Reader

>Midjourney is more popular (from what I'm seeing) than Stable Diffusion at the moment because it's better at the moment. Midjourney is closed-source.

Midjourney is easier, its not better. The low barrier to entry has it popular, but it isnt as realistic, doesnt follow the prompt as well, and has almost no customization.

SD is the holy grail of AI art, if you can afford a computer or server to run SD + have the ability to figure out how to install python, clone Automatic1111 from git, and run the installer, its the best. Those 3 steps are too much for most people, so they default to something more like an app. Maybe it is too soon, but it seems SD has already won. MJ is like using MS paint, where SD is like photoshop.

View on HN

I just want to add my $0.02 currently working at a games studio that is integrating AI generated art into our art pipelines.

Midjourney definitely generates really high quality art based on simple prompts, but the inability to really customize the output basically kills its utility.

We heavily use Stable Diffusion with specific models and ControlNet to get customizable and consistent results. Our artists also need to extensively tweak and post-process the output, and re-run it again in Stable Diffusion.

This entire workflow is definitely beyond a Discord-based interface to say the least.

If you would give a talk about this, I would watch it - despite being out of the graphics for almost 10 years now. Really want to hear from the trenches about the workflows, benefits and challenges you have.

Here is a test I did the other day of rough sketch (hand drawn) -> clean line work (AI) -> coloured (AI). This workflow gives 100% control over the output because you can easily adjust the linework in the intermediary step.

https://twitter.com/P_Galbraith/status/1649317290926825473?c...

This is using Stable Diffusion and the Control Net Lineart Model. The coloured version is pretty rough but it was a quick test.

In my opinion Stable Diffusion is vastly superior to Midjourney if you have the skill to provide input to img2img/ControlNet.

I have some other earlier workflow experiments on Youtube if you're interested in this kind of thing https://www.youtube.com/pjgalbraith

use https://github.com/deep-floyd/IF, it uses LLM to generate exact art you need.

The image quality of DeepFloyd is much lower than Stable Diffusion 1.5 though, it's a pretty major tradeoff. Can definitely be part of the workflow since it really is good at composition, but right now it's not a replacement.

Deep Floyd doesn't allow commercial usage, such as a game studio using it.

They said that they will change the licence for the final version IIRC.

> AI generated art into our art pipelines

I'd be interested to know where the art ends up in the game? Do you mean 2D backgrounds and billboards in-game? Or are we talking cut-scenes and menu screen art?

Really curious about the challenges you've been seeing as your art team's been trying to integrate AI generated art into your pipeline.

Excited by the potential for artists to accelerate their creative process with advances like ControlNet, but makes sense there's a lot frustrating about the process today.

Exploring tools that might help. If you might be interested in chatting, email me at tom@adobe.com. Thanks!

> Midjourney is easier, its not better.

By what measure? Midjourney v5 is massively better with every prompt topic I've thrown at it than SD. It's not even close. SD, however, is much better if you want an actual customizable toolchain or to do things like train it on your own face/character.

They also just released v5.1, which seems to be quite a bit better than v5.

Is there a comparison? It is interesting that there do not seem to be any Midjourney benchmarks. E.g.

https://paperswithcode.com/sota/text-to-image-generation-on-...

Parti and Imagen are still on top, followed by Dall-E 2.

If their model is so great, why are they afraid of benchmarks?

I literally just don't feel like running them tbh, and see no reason to publish them either way. Mostly prefer to let the outputs speak for themselves.

For a while I was using an FID variant for evaluation during training, but didn't find it very helpful vs just looking at output images.

Okay. That's probably the difference between a commercial and a research project.

Agreed. I pay for MJ and have several SD versions running on my PC. I like the ability to fine tune the SD models and my Pc with a 4090 is plenty fast, but I can't match MJ's output on artistic quality. SD allows for 4k sized outputs which is great but I can't use the art like I would like. FWIW the SD NSFW community is large but that is not where I invest my time with AI art.

Generate the following picture in mid journey: "A school of dolphins spanking a mermaid with their flukes."

A 1000 V-rolls won't get you there. For something like this control net combined with inpainting is indispensable. Not to mention the excessively heavy handed censorship in MJ.

Midjourney excels in overall quality, but it completely falls down if you have an actual complex vision.

It seems Midjourney is great at generating non-pornographic pictures.

Uhh... Spanking isn't inherently pornographic - that particular image was supposed to be a Gary Larson parody style comic.

Here have another prompt: "Rapunzel has let her hair all the way down a tower. The hero has been tied up by the witch, and annoyed at Rapunzel's continual attempts to escape, the witch throws the bottom of her hair into a paper shredder at the base of the tower."

You could v-roll until the heat death of the universe without even getting close.

Midjourney is great if all you're capable of conceiving is 90s Mad Magazine templatized mad Libs, banal crap like "Darth Vader as a French pantomime street artist".

Unfortunately that also describes the majority of midjourney users.

Yeah, prompts which basically just list properties (Darth Vader, French, pantomime, street artist) seem to work well, but relations are mostly too hard for these models. Even "a monk playing chess against a clown" or "a blue book on top of a yellow book" is out of reach for Bing's Dall-E ~3, and Midjourney probably isn't much better here.

https://www.bing.com/images/create/a-monk-playing-chess-agai...

https://www.bing.com/images/create/a-blue-book-on-top-of-a-y...

Simple (prompt only) use of generative models is quite good at creating simple artistic pictures you might actually hang on a wall. Trying to create a complex scene with just a prompt seems still a few years off though.

That seems to depend on your use case. Frankly, I don't have much use for either of them but Midjourney was much closer.

I've twice spent a couple of hours unsuccessfully trying to generate a simple background image that would be blurred out when rendering 3D models. SD out-of-the-box was far worse, but Midjourney still was not up to the task. It's incredible how well they can generate images of nearly any subject/object and make some changes to the style and placement, but trying to precisely achieve critical broad-stroke things like like perspective, sizing, lighting direction/amount/temperature, etc. was far too cumbersome. Prompt refining is just like having a program with a bunch of nebulous undocumented menu entries that you just have to click on to see what they do rather than just giving you the tools to make what you need to make. Was that the right entry or the wrong entry? Who knows! Maybe just try it again to see if it works better!

There's a fundamental disconnect between professional-level and consumer-level tools. Consumer tools must be approachable, easy to use, quickly yield appealing results, affordable, and require little maintenance. Professional tools need to be precise, reliable, capable of repeated results with the most demanding tasks, and easily serviceable into perfect working order.

These are consumer-level tools. If you merely need a cool picture of a space ship done in such and such style with such and such guns blah blah blah (that for some reason always looks 10%-50% Thomas Kinkaid,) these tools are great, but they abstract away the controls that really matter in professional work. Novices who get overwhelmed by all of those factors love it because they don't understand, and probably don't care about what they're giving up. For serious work, aside from getting inspo images or maybe generating deliberately weird bits of whatever, they're hit-or-miss at best. Without exception, doing a rough mock-up in a modelling program took FAR less time than trying to wrangle exactly what I needed from one of those generators.

I'm sure they'll get there someday but right now they're miles away from being professional-quality image generation tools.

I will say though that low-effort higher-volume professionals (e.g. mobile game mills, Fiverrrr designers) will likely profit from these tools once they can out-compete cheap online assets from stock images/models/etc. but they're so not there yet.

> Without exception, doing a rough mock-up in a modelling program took FAR less time than trying to wrangle exactly what I needed from one of those generators.

I think a lot of people have unrealistic expectations of the tech-- they think they can get exactly what they want if they are articulate enough in describing it with words.

Feed your rough mock-up to img2img (or use inpaint sketch) and you'll land much closer to where you're trying to go.

It's a power tool. It will do tedious manual work (producing art) very quickly. The difference between professionals and consumers in how they use it is that the professional asks the machine to "finish what I started," whereas the consumer tells the machine to "do all of the work for me."

I tried img2img. It will do rough finishing work but it won't take some lighting vectors and match my lighting. It won't shift the viewpoint by 18 degrees. It puts a smooth sheen on rough work with broad stroke needs and that's valuable in some cases, but it is not a general-purpose professional tool.

Canva competently satisfies most non-professional needs but it only satisfies a narrow slice of professional needs. Trying to use it for most professional work takes vastly more time and effort than using a proper professional tool. LaTeX fits academic paper publisher's needs and can pump out formatted and laid-out text far quicker than someone using InDesign but you'd go crazy trying to assemble a modern magazine or high-end book. It doesn't need polish or sheen. It needs something fundamentally structurally different.

I'm both a professional digital artists and a long time back-end software developer. This slice of time has really opened my eyes to what it must be like for most non-developers to speak to developers: constantly oversimplifying your use case and assuming some algorithmic approximation will do without really understanding the problem.

Fair enough. Your 3D modeling needs might be a bit advanced for the current state of things. It works pretty well for flat graphic design, stock photo or illustration purposes.

I'm holding out for an instruct-based model that will take explicit instructions, or at least layered prompts. Mutating the original prompt along with the picture (or changing the entire thing to only describe certain parts, a la inpainting) is frustrating to me.

That's just one specific example of why it fails as a high-level general professional tool. Even for tasks like straight-up photo finishing... might be OK for making some neat thing or adding some generated detail to a phone shot but no way that's going to finish someone's professionally shot photos. It might replace Fiverr graphic design that was probably done with a template but real graphic design has conceptual meaning. Making the actual assets is the easy part.

I think what most developers don't realize is that creating media, even snazzy, captivating images, is a tiny portion of the work commercial artists and designers do beyond Fiverr or the people working at sign shops. We've got bazillions of stock photos, asset stores, etc. etc. etc. available at our fingertips at a pretty low cost... and we use them, just as we'll use AI generated images in our processes... Just not for anything that actually matters. Reducing art and design to what so many people have is nearly akin to reducing coding to how fast someone can generate code regardless of it's suitability for the purpose. I could go into Adobe Illustrator and make most of the vector art you see on the net in a few minutes. Knowing what needs to go on the screen, exactly how it needs to go on the screen, and the effects of putting it there, what it communicates and to whom, etc. etc. etc. are the hard parts. You could have a tool that would instantly translate millions of pretty images from someone's imagination in a matter of milliseconds but that's not going to make them useful for communication or anything else.

So very much of this hype is people simply assuming that they understand something that they don't. And that's why the UIs for most FOSS applications suck yet the project contributors will defend them like their own children.

I think a lot of people have unrealistic expectations of the tech-- they think they can get exactly what they want if they are articulate enough in describing it with words

Who's fault is this though? The hype is absolutely hysterical.

Thanks! That's pretty neat. The sample images look a bit overwrought like a lot of other AI images do but I'll bet they're doing that to follow the trend rather than it being a technical limitation.

Wrong thread?

yep. Really need to stop commenting when I'm waiting on after-hours compiles.

ControlNet helps a lot with composition and lighting (https://sandner.art/create-atmospheric-effects-in-stable-dif...). It requires more work than just entering a prompt, but probably less work than doing it manually once you get used to it. I think there's a number of StableDiffusion clients in development that are trying to make this easier.

Ha... I accidentally replied to the wrong comment. Anyway, thanks! That's pretty neat. The sample images look a bit overwrought like a lot of other AI images do but I'll bet they're doing that to follow the trend rather than it being a technical limitation.

Classical professional tools like Photoshop have a lot less potential though. They are very precise, but (I assume) they have barely advanced in the past decade. Tools based on generative AI will probably improve massively over the next few years. Most such tools seem currently based on Stable Diffusion, and apparently OpenAI/Midjourney/Google have zero interest in supporting such tools. But this could change soon, e.g. when Adobe tries to compete with the SD ecosystem.

We already now see deepfakes (e.g. of Trump or the Pope, recently even videos) that a far beyond what we saw in the years before, indicating that the old professional tools weren't so powerful after all. Now if we extrapolate this a few years into the future...

You assume wrong. What these tools offer is constantly churning. Far faster than ever before and Photoshop has been around for over 30 years. They release updates constantly. Photoshop got AI filters like detail enhancement for zooming a few years ago. Automatic object detection, content-aware delete, etc. etc etc. a few years before that. That's only what I can recall off the top of my head for Photoshop alone, but it's such a giant environment that even most of their own product people probably couldn't tell you off the cuff. In areas like video compositing, tools like Nuke are developing tools with these capabilities even more quickly... and they better when the cheap license costs $3500/yr.

As I mentioned in another comment, so much of this hype is based on developers assuming they understand something that they don't. I've indulged in this hubris as a developer but straddling both sides of this line has been illuminating.

Well, Adobe at least seems headed to fully embrace the generative AI hype now:

https://www.adobe.com/sensei/generative-ai/firefly.html

This sounds all very similar to the Stable Diffusion tool chain, though probably cloud based and with a more intuitive UI on top.

All of these technologies are being integrated into professional toolkits in ways that make sense when they're polished enough to be professionally useful... and for the foreseeable future, that's how it will stay. Beyond the high-volume low-effort work on places like Fiverr, commercial artists and designers are valuable for their ability to think conceptually and make the artistic decisions about what goes on the screen, where, and why. The how is an implementation detail. Designers dropped balsamiq in favor of sketch in no time flat, and then dropped sketch in favor of figma even more quickly. Adobe XD, capable and included for free in an ecosystem they already use, it's barely in the conversation. These are fields where people readily adopt new technology that suits their needs but the current tools aren't even in the ballpark.

Being able to quickly generate and iterate on assets is great for inspiration but pretty useless for professional output without fine-tuned, predictable, repeatable controls. These tools will simply integrate with existing professional tools until they can do it better.

Imagine the first person to make an electric saw made some automated thing that could cut the wood to make a cool looking flat pack house somewhere in the neighborhood of your specifications in 5 minutes. The caveats: while it would assemble perfectly, the actual angles of the cuts might be unpredictable... Like 40 and 50 degrees rather than 45 and 45, and the layout was never quite what you expect even if it was OK more often than not. Pros knew those were fundamentally deal breakers for professional work, and remained more professionally useful with their hand saws because they had the required precision, control, and predictability. While enthusiasts were gong crazy exploring all of the different kinds of oh-so-slightly wonky structures they could generate and predicting the end of carpentry, the old school saw companies started making circular saws, chop saws, drills, and the like. The market for handyman-built dog houses, sheds and playhouses would immediately be lost to the automated machine but I guarantee you that all consequential work would still be done by carpenters with power tools.

I agree that for the foreseeable future designers / commerical artists will have to overall decide "what goes on the screen, where, and why", but not anymore for all smaller scale details. Generative AI automates creativity at least to some extent, which is qualitatively different from past developments. The example with the island tortoise on the Adobe website is not a real demo, but it doesn't seem far away.

I have SD up on a machine with a 3090 and it can't produce output half as good as MJ without a ton of work.

I use SD to augment MJ, like fixing hands with specific LORAs for example, so I definitely appreciate that it exists. But for actually creating a full image in one shot, they're not even comparable.

> Midjourney is easier, its not better

Does being easier not influence whether it's better? I mean this in that for many of the ways AI art would be used, MJ already seems to be "good enough" at a lot of it.

Secondarily: doesn't Midjourney's increased user base and increased ratings they get from users help it refine its model, thus meaning that "ease of use" creates a feedback loop with "quality of output" because more users are engaged?

I'm asking real questions, not making a statement I believe in and just adding a question mark.

>Does being easier not influence whether it's better?

As mentioned, what is better, MS Paint or photoshop? If MJ ignores your prompt and spits out a half related picture, are you going to continue using it?

If anything MJ is a stepping stone to SD. You get a taste of AI art, but want to do something specific that MJ cannot do. You learn about control-net, alternative models, inpainting, etc... and you decide you need to move on from MS Paint to Photoshop.

I personally used free AI art(cant remember which), it was super cool, but quickly I wanted to use different models and generate thousands of pictures at a time. I wanted to make gifs, img2img, etc... and the only people doing that were on SD.

I think you actually have that backwards, because your conception of "easier" is a bit skewed.

The question is not "which is easier?", but rather "which is easier to use to produce high-quality output". In your analogy, I'd argue the answer to that question is actually Photoshop. Likewise the answer in the MJ/SD case is MJ.

> As mentioned, what is better, MS Paint or photoshop?

Metaphor is useful, but this feels overly-reductive. The gap between the amount of effort it takes to make something great or approaching the vision you had is massive between MS Paint and Photoshop. Not so for SD and MJ.

However, I am appreciating that SD seems to be clearly better if you need something more specific / precise. I don't think I'm convinced (yet) that because it can get more precise inherently makes it a better tool.

It's the old consumer/professional distinction at play: "If it's a professional tool, it's a job to know how to use it."

There are definitely some professionalized paradigms emerging in the use of SD: one video tutorial I saw this morning covering basic photobash + img2img and use of controlnet had a commenter saying that they preferred using the lineart function in controlnet to get more control and leverage their drawing skills.

When you see that kind of thing it's a huge signal for professionalization, because someone suggesting the optimal first step is "learn how to draw" deviates so completely from the original context of prompt-based image generation: "just type some words and the image magically appears".

What is better oils or acrylics? What is better, clay or wood when it comes to sculpture.

It is more like comparing Krita vs Photoshop than MS Paint vs Photoshop. That is bogus.

Most AI art I have seen is complete shit anyway and especially from SD.

> Does being easier not influence whether it's better

Midjourney let's you type in a thing and get a result that will look great, which is no small accomplishment. If you want "incidental art" like blog post heros there is no competition. But it's really hard to use if you want to get exactly what you want.

> But it's really hard to use if you want to get exactly what you want.

Alternatively, if you don't know what you want, it's really good for inspiration.

Very true, I've fleshed out RPG scenes/characters with it.

> Does being easier not influence whether it's better? I mean this in that for many of the ways AI art would be used, MJ already seems to be "good enough" at a lot of it. > ... > I'm asking real questions, not making a statement I believe in and just adding a question mark.

In my opinion, it is tedious to discuss whether Midjourney or Stable Diffusion is better.

Both are tools with their own strengths and weaknesses.

Midjourney

- Comparable to a ready-made consumer product.

- Easy to use and you get excellent results quickly.

- The feature set is limited. So you can't influence so well how the result should look like.

- Restrictions on what you can generate with it (gore, nudes).

- Fee required.

Stable Diffusion

- Comparable to a development kit. Technical, Difficult to use, different tools that you have to configure and use correctly.

- For excellent results you have to invest much more time.

- Extensive feature set (inpaining, outpaining, upscaling, ControlNet, many different models, LORAs, textual inversions). You can much better define how the result should look like.

- No restrictions on what you can generate.

- Free if you install it locally.

Midjourney is higher quality by a fair bit, from my personal experience and from being near a few of the top early AI artists for a good little while.

Is midjourneys model actually better?

I was under the impression that midjourney was just running a form of SD and it's real secret sauce are the peripheral prompts it injects on the backend along with your prompts.

I could be totally off the mark here.

To me it is like saying oil is better than acrylics. These statements have no meaning when it comes to art.

The model is obviously massively better. and they haven't been using SD in any form since the test mode of v3. the models are trained from scratch.

Is midjourney still using discord as its primary user interface? That really turned me off.

Yes-- a classmate uses it. They do @everyone announces in their server every day, and while you can mute actual notifications, it still adds one to your badge count. My attention is too valuable-- that would get me to cancel my subscription.

When I used it, it didn't feel like a product. It felt like a demo. If I have to rely on some bot in a public forum, I'm not sure I like this product.

I've operated on the assumption that MidJourney is deliberately knee-capping their growth by making it only work on Discord to ensure they don't grow faster than they can add hardware.

I could be entirely wrong, though. Maybe the person making that decision is just an idiot.

I have an IRC bot that has triggers for DALL-E, GPT3, and ChatGPT. I really want to make one for MidJourney, and I would happily pay MJ for the privilege. But I can't. Not without breaking some rules.

They have a <10 person team IIRC, using Discord as an interface saves a TON of money/effort/maintenance/risk, etc, as best as I understand, and lets them focus on the technical product. Remember, Discord is giving them free content legal protection by proxy, even if that's not necessarily the original intended effect I think. There's a lot to gain by riding alongside Discord as a primary interface vehicle, I personally believe.

They're good enough technically at what they do that their audience is okay with the interface that they have to use, I'd reckon (I've heard similar beefs about the UI stuff though, so it sorta makes sense).

> Discord is giving them free content legal protection by proxy

Would you mind explaining what you mean by that? Thanks!

Discord does not have its own data centres, It would be hard for them to provide such a guarantee.

Discord hosts a substantial number of images among other things.

Even if you turn all permissions on for your bot including the ability to use slash commands it won't be able to send commands to the Discord mid journey bot. (And there is no publicly exposed API)

The only way to do it right now would be to create a separate dedicated discord account linked to midjourney, and then have your bot control that user account and that's a good way not only to get your MJ access revoked but also to have your discord account banned.

As far as kneecaping their growth, they have one of the largest discord channels in history and they frequently run up against compute limits, so if this was one of their ostensible goals they failed at it pretty spectacularly.

> Even if you turn all permissions on for your bot including the ability to use slash commands

AFAIK, this isn't possible.

MJ edits your prompts, you can achieve the same level of quality if you use the same prompts they do (which can be found on the internet)

They edit your prompt ... how?

By adding keywords to the prompt and to the negative prompt

Can you give me an example? What are those keywords? This sounds a bit like Dall-E 2, where OpenAI appears to manipulate certain queries involving women to make them black.

https://www.reddit.com/r/StableDiffusion/comments/y649yn/pro...

Nothing so… social

Okay, though this seems to be someone who tries to imitate Midjourney using Stable Diffusion, so it is still not quite clear how Midjourney does it. Though it seems plausible that they would "cheat" in this matter in order to get something more artsy looking. (But what if you don't want something artsy, or not artsy in their style?)

Thats the value proposition of MJ I guess. Its easier to get something good out of a simple prompt, but you end up with a recognizable MJ look

Are you sure about this? For the couple things I tried, a colleague with Midjourney managed to outperform my attempts with SD by leaps and bounds.

There's a higher "skill ceiling" with SD. You can install different models for different styles or subjects, use ControlNet for composition, and use plugins to do things you can't easily do with MJ.

> SD is the holy grail of AI art, if you can afford a computer or server to run SD + have the ability to figure out how to install python, clone Automatic1111 from git, and run the installer, its the best.

If you can afford Colab (which is free if you don’t want to use it too much), you can just click one of the existing A1111 colabs and run that, you don’t need to figure out python, git, or A1111 installs.

Free Colabs have started blocking any SD web-ui it detects (presumably because it's meant as a community service for ML researchers, not for people who want to play hentai gacha, and they're running out of server time)

Google is cracking down on this recently.

Cracking down on what exactly?

TLDR here: https://twitter.com/thechrisperry/status/1649189902079381505

Do you have a link to a decent tutorial for someone to do what you're describing in the last paragraph?

Do you need additional detail that cannot be found here?

https://github.com/AUTOMATIC1111/stable-diffusion-webui

Or are you looking for the cutting edge stuff like control net?

If you want to use colab instead, I used this a month or two ago.

https://colab.research.google.com/github/TheLastBen/fast-sta...

I hope other people can give you further reading.

There are dozens on YouTube. My kids did it, and they don't even program and had never touched Python in their life.

Even trained their own models using a cloud GPU.

The SD ecosystem is wild.

It took me a little hunting, but thanks to Reddit I eventually found a cloud-gpu host that provides a working Stable Diffusion image. So you basically don't have to do anything that GP said. Everything is installed and you just rent the hardware.

https://www.runpod.io/console/templates

Look for "RunPod Stable Diffusion". I spent a whole $0.35/hr playing around with my own SD instance that I had running in minutes.

35c/hr seems crazy expensive compared to Midjourney. Midjourney gives you set fast hours (Immediate GPU) and unlimited relaxed hours (Delayed GPU). It also has a lot of built-in parameters you can use to easily tweak images. I'd rather pay for MJ than run my own SD.

The main upside of running your own SD is that you can completely automate it, but I'm not sure how useful that really is.

> The main upside of running your own SD is that you can completely automate it

No, the main upside of running your own SD web UI is that you can select and deploy your own checkpoints (not just using the base SD models), LoRas, embeddings, upscaling models, and UI plugins supporting additional services/models/features like multidiffusion (bigger gens and controls of which areas within the image different prompts apply to), ControlNet and associated models, video synthesis, combinatorial prompts, prompt shifting during generation to do blending effects, and, well, a million other things.

Also, you can completely automate it.

The midjourney price would be equivalent to ~100 hours cloud time. How is that crazy expensive?

You can do the same thing on vast.ai too.

It's a little inconvenient to use non-base models and plugins this way (you pay extra for more storage), but it's definitely an easy way to use the full power of SD.

https://rentry.org/sdg-link

Midjourney retrains itself, I have one click installer apps for SD, Midjourney and the live prompt community is very good

None of this stuff is copyrightable so I dont care that its not private

Midjourney is more niche. It’s great at photographs, digital art, concept art, game art and everything in that sphere. Because that’s what it was trained on. So it has a specific style. Dall-E in comparison produces kind of garbage looking pictures of many more styles

Reply on news.ycombinator.com