Some time back, I rendered some landscape paintings in the style of Casilear. I don’t know whether the bot can do different-resolution images — I did those at 2560x1440, whereas the bot is using 1024x1024 in what I’ve seen, and I don’t see a reference in a quick search of the docs to a way to change that. I was also using a model, realmixXL_v15, derived from Stable Diffusion XL but not itself Stable Diffusion, and it doesn’t look like AI Horde has that model available. So I don’t know how well this will work. But let’s see what happens if we ask it to regenerate such an image using SDXL:

@aihorde@lemmy.dbzer0.com draw for me Scotts Bluff, by John William Casilear style: sdxl

Man, that ran quickly enough that whatever rendered the thing had to have had Pony Diffusion either already downloaded or I’m misunderstanding how AI Horde functions…maybe it normally farms out images to the cloud and is able to intelligently try to find a node that already has a given model downloaded rather than reloading a new model or something.

Okay, a repeat, using Nova Furry Pony:

@aihorde@lemmy.dbzer0.com draw for me score_9, score_8_up, score_7_up, cat_girl, business_suit, morning, drinking coffee style: nova furry pony

Now, let’s try out another model, just to make sure that styles that entail having a model downloaded to a node actually work with this. Looking at the bot’s history, it looks like it has been used to render things in other styles, but I don’t know for sure whether those are styles that involve use of another model.

Pony Diffusion doesn’t really take English-language prompts (though it’ll probably do something not-entirely-unreasonable with them); it really expects a list of keywords. However, it does understand Danbooru tag groups:

https://danbooru.donmai.us/wiki_pages/tag_groups

@aihorde@lemmy.dbzer0.com draw for me score_9, score_8_up, score_7_up, cat_girl, business_suit, morning, drinking coffee style: pony

Yes, the default style is “flux”. Excellent!

Now I just want to confirm that the default style is Flux; I think it is, from db0’s comments. That would be good, because Flux is pretty easy to use and can just take what more-or-less amounts to English-language prompts. Many other models want something more like a list of keywords.

@aihorde@lemmy.dbzer0.com draw for me A delicious slice of cheesecake with a delicious scoop of strawberry ice cream on the side and chocolate shavings sprinkled on top. The cheesecake is on a white plate.

Okay, it differs, so that probably means that it’s set up to use a random seed. Hmm. That’s not my preference, since it makes it hard to narrow down and refine a precise image, but some people seem to like that as a way to generate more images.

To repeat the first test post I ran (just to see whether it gives the same output, which would mean that it uses a fixed seed, or whether it gives different output, which means that it uses some non-deterministic process):

@aihorde@lemmy.dbzer0.com draw for me An engraving of a skunk. style: flux

Hahaha, that’s fantastic. Thanks, @db@lemmy.dbzer0.com. I can’t believe that I had been unaware of this for years.

I’ve got local hardware myself, but this makes generation way more accessible to random users on the Threadiverse without tying them to commercial services, including users who are using a phone and can’t really run the stuff locally.

skims bot FAQ more

It does look like the bot tries to block NSFW stuff, but outside of that, it looks pretty permissive of whatever, and while the style list doesn’t clearly distinguish between styles consisting of prompts for one model and different models, there are a lot of models in there that I recognize. Just append style: flux to the end of the prompt. It’s got Flux (style: flux), Pony (style: pony), Pony Realism (style: pony realism), the Nova family (style: nova anime xl, style: nova, style: nova furry pony), Stable Diffusion 3 (style: sd3), SDXL (style: sdxl), Mistoon (style: mistoon anime), just from a very quick skim. I guess AI Horde must have a list of models and the ability to just download models or something if it doesn’t already have them present on a node?

skims bot FAQ

Oh, dang, it even runs multiple models, and it looks like it can run Flux, which can handle fairly English-language-looking prompts.

I gotta see this.

@aihorde@lemmy.dbzer0.com draw for me An engraving of a skunk. style: flux

Hahaha, awesome, db0, didn’t know that you’d set this up until now.

1986 – The Mindbender derails and kills three riders at the Fantasyland (known today as Galaxyland) indoor amusement park at West Edmonton Mall in Edmonton, Alberta.[10]

Huh.

https://en.wikipedia.org/wiki/Mindbender_(Galaxyland)

On January 30, 2023, the mall decommissioned and closed the Mindbender after 37 years of service, in order to redevelop its space for new developments in the park.[4] Its trains were reused for All American Triple Loop, at Indiana Beach, United States.

https://en.wikipedia.org/wiki/All_American_Triple_Loop

All American Triple Loop (formerly Montaña Infinitum [“Infinity Mountain”] (2007–2014), Montaña Triple Loop [“Triple Loop Mountain”] (2014–2016) and Quimera[1] [“Chimera”][2] (2017–2019)) is a steel roller coaster at Indiana Beach in Monticello, Indiana.

Manufactured by Anton Schwarzkopf, it was originally purchased by showman Rudolf Barth in 1984 who operated it as Dreier Looping for 12 years on the German fair circuit.

After this, it was the main attraction in three major theme parks: first spending 2 years in Sunway Lagoon as Triple Loop Coaster, next, it spent 5 years in Flamingo Land resort as Magnum Force, and finally at its third and most recent location at La Feria Chapultepec Mágico, as Montaña Triple Loop. In 2017 it was renamed Quimera. In 2024, it opened at Indiana Beach as All American Triple Loop.

I feel like a secondhand German roller coaster that went from Germany to Malaysia to England to Mexico now running secondhand Canadian trains arguably isn’t best named the “All American Triple Loop”.

The review to determine whether the US should scrap the project is being led by Elbridge Colby, a top defence department official who previously expressed scepticism about Aukus, according to six people familiar with the matter.

Eh. Looking back, it sounds like his take is fairly nuanced. From what he’s said in the past, it sounds like he’s said that he doesn’t think that entering the arrangement was actually worthwhile for the US, but that he’s also hesitant to withdraw from an agreement once entered into.

https://thenightly.com.au/politics/world/elbridge-colby-man-vying-to-be-donald-trumps-next-security-adviser-questions-viability-of-aukus-c-15058991

It sounds like his argument is that the main risk of a military conflict with China over Taiwan, where these would play a role, is relatively near-term. Australia hasn’t stated that it would defend Taiwan, and AUKUS won’t result in an aggregate increase in submarines across the US and Australia for some time, which means that it would reduce the number of submarines available to fight China.

Assuming that all that is accurate, that seems to be a fair take to me. My guess is that what he’s actually after, given his phrasing, is not trying to trying to end AUKUS, but to get Australia to also commit to defending Taiwan as a condition for it.

Speaking in London on Monday, Mr Colby said that US shipbuilding could not keep pace with the target of delivering Australia subs by 2032 and questioned why the US was giving away its most lethal assets to a country that was not even guaranteeing it would use them in the event of a conflict over Taiwan.

“If I were king for a day on the subject I would say ‘Look, you all know what my concerns are, let’s see if we can work through these together’.

I’ve no idea whether that’s something that would be totally unacceptable to Australia or not.

At one point, the US did try to put together an analog to NATO in the Pacific, SEATO. It didn’t really go anywhere. But conditions have also changed.

https://en.wikipedia.org/wiki/Southeast_Asia_Treaty_Organization

Australia was a member. Taiwan was not. But the US might be aiming to build a new Pacific alliance today.

I think that that’s pretty much a subjective question. I could tell you what I find tacky, but it won’t be a universal. Some people will have different positions, and there isn’t one “correct” view to have. It’s like asking what the best flavor of ice cream is, or the best song.

https://en.wikipedia.org/wiki/White_House_Rose_Garden

I think that’s gravel, looking closely at the image.

That being said, I don’t know whether concrete walkways were much of a thing in 1908. Gravel might have been the closest common option to what we’d more-commonly use concrete for today.

Not as big a deal in DC’s climate, but here in often-arid California, it’s generally considered responsible to phase out water-hungry lawns in favor of landscaping that doesn’t require as much water. Drought-tolerant plants, gravel, rock gardens, concrete, whatever.

https://en.wikipedia.org/wiki/Xeriscaping

The lawn was an English custom, and trying to reproduce a little piece of wet England by pouring enough water on arid or semi-arid land day in and day out is kind of wasteful.

If you need the cushiony and rapidly-self-healing properties of lawn because people are running around on it, that’s one thing, but people spend more time indoors than they did historically, and as just a thing to look at, it’s not a great default. Plus, kind of high maintenance.

A factor for a number of Western states.

That being said, this wasn’t the rationale, and frankly, it’s probably basically irrelevant for somewhere like the White House relative to the functional impact.

The woman in the painting is Anna Christina Olson (May 3, 1893 – January 27, 1968). Anna had a degenerative muscular disorder, possibly polio or Charcot-Marie-Tooth disorder, which left her unable to walk.[1] She was firmly against using a wheelchair, so she would crawl everywhere. Wyeth was inspired to create the painting when he saw her crawling across a field while he was watching from a window in the house.

:-(

Man, I’m glad that we figured out the polio vaccine.

A couple of my favorites from that thread:

@FlyingSquid@lemmy.world

@ThunderWhiskers@lemmy.world

@YodaDaCoda@sh.itjust.works

@DredUnicorn@lemmy.world

@Thelsim@sh.itjust.works

@anonymoose@lemmy.ca

@itslilith@lemmy.blahaj.zone

@starcat@lemmy.world

@Tlaloc_Temporal@lemmy.ca

@Usernameblankface@lemmy.world (also OP here, also the person who brought the idea up in another thread)

@WandFliesenWodka@lemmy.world

Link to the original version of this thread to which OP is referring, which was quite a hit some time back:

https://lemmy.world/post/7371120