1. Download https://addons.mozilla.org/... (or whatever the chrome link for it is)
2. Open the popup
3. Click the lemmynsfw.com part of the address under Write style for (otherwise it'll use the full URL and won't apply to the all the site)
4. Paste this in:

.img-blur {
  filter: none;
  -webkit-filter: none;
  -moz-filter: none;
  -o-filter: none;
  -ms-filter: none;
}

Yeah, with AI you really have to work towards actively telling it not to generate that kind of stuff, and they slip through the cracks regardless. Report the ones you want gone and I'll remove 'em (even if it's one I posted)

By Web UI it means that the graphical part of it -- where you write your prompt and hit generate -- is running inside your browser and not as a separate window or command line. Everything is kept on your own computer unless you explicitly tell it to open up remote access.

Report them and we'll remove them. This community is starting to get a bit more active than I anticipated so I can't really look at every post by myself like I could at the start.

If we leave the reposting to lemmit, people can easily opt out from it showing up on their All feeds by blocking the singular bot that does all the posts, whereas if we allow it here people would need to individually block each bot.

However, Lemmit only reposts Reddit, there is at least one bot here that reposts one of the rule34 boorus, and I'm sure other sites are to follow. Those seem to be the real issues here.

Reddit doesn't own that content, individual posters do. They do grant Reddit a license to display it to others (in the terms of service nobody reads) but they are still the legal owners of their posts (assuming we're talking about OC)

Someone had actually DM'd me ahead of time asking about it and I never realized I actually said this publicly until recently.

I mean, as long as you have consent (and considering you're talking about yourself I think it's fair to say you do) it should be fine.

I explicitly mention this on the sidebar (which is where the "canonical" version of the rules are), but since most people prefer generating celebrities and other people who definitely have not consented I did decide to phrase it a bit more broadly for this announcement.

ROCm is flaky in regular consumer GPUs at the best of times. I'm surprised you could even get that far on a steam deck.

Try the command line arg --opt-sdp-attention. You might also want to try out --medvram or --lowvram (4GBs is considered low when it comes to AI). although I have a feeling it's just because of the custom nature of the deck's APU.

Your best bet would be to search for builds of ROCm, PyTorch and Torchvision that are specifically made for the deck, if such things even exist.

For kickstarting I personally just dumped the relevant ones from my own stash (that were still up, I have a few more I'll need to reupload because Imgur & co) and let the community take over from there, though of the 3 communities I tried it only seems to have worked for 1, so YMMV.

!orgasm@lemmynsfw.com

It's probably "Add python to environment variables." They must have changed the wording on that at some point.

how much space was the download(s)?

On my end, it's sitting at ~64GB (with btrfs compression shenanigans), though 60 of those are from all the models I have installed. The download would probably be ~2GB, even less if you disable downloading the "default" models with --no-download-sd-model and instead pick models off of Civit or wherever manually.

Edit: Should have mentioned. Most full models are between 2-4 GBs each. Some can be 5+ but they tend to be "full" versions intended for merging & such. LoRAs are generally smaller. Depending on how much they're pruned they'll be anywhere between 10-100 MBs each.

how confusing is the software to use?

There's definitely a learning curve, yes. But there's plenty of resources (and more importantly, examples) out there.

what kind of limitations does the software have? can i do multiple people? monsters? futa? etc.

As long as you have the correct models set up it can generate basically anything. At least with anime models, monsters and futa are a given. Your main issue will probably be multiple people, although there are solutions to that. (See the multidiffusion upscaler GitHub repo on the main post)

Best of luck, both to you and whatever you decide to do in the future, but also to the poor soul who will now have to maintain all this after you :p

It's all free, yes.

For the how: Aside from the webui (or whatever else you're running the stable diffusion code on), you'll need models that instruct it what to create. The builtin models are terrible for porn.

I think what's happening is that the new auto-NSFW patch is checking the box visually, but isn't letting Lemmy know about the box actually being checked.

Here is the code that seems to be behind it. It's setting .checked true, but there's a change event Lemmy registers on the checkbox, which I'm not entirely sure if it gets triggered or not.

This seems like a really roundabout way of going with this, but I'm not sure how hard it would be to do the 'proper' way. It's 4 AM and I really should be sleeping instead of lurking here.

If you're sure you enabled Show NSFW in your settings, it could just be that your instance didn't want to backfill older posts. No idea if it tries to do so when you visit a previously unvisited community for the first time.

Edit: Of course if you're visitng without being logged in, it won't show anything. No idea why Lemmy does it that way.

Half my time here's spent looking for un-NSFW'd posts in my communities (specifically this one, nobody posts in the other) and telling people to NSFW their posts. At this point I don't think I actually fapped to anything in ths entire instance yet.

The admins are trying their best getting the situation under control by modifying Lemmy, but as mods all we can do is to remind posters when they forget. And the Lemmy devs seem to have shot down the idea of giving mods edit rights. You can go through my comment history for the link to the GitHub issue.

I don't think that's necessary. Not sure though.

Not on Windows right now so I can't confirm, but you probably forgot to pick "Add Python to PATH" or whatever the option is in the installer. Try running the Python installer again, maybe it'll let you add it without needing to uninstall & reinstall

Edit: If you're on Nvidia, there seems to be a simpler install method now: https://github.com/... (Method 1)