Is that your boyfriend or something?

You have to understand that there's not a lot of people who work on the Perchance website so it's going to take time for stuff to get taken down. I'd heavily suggest you get in contact with an admin or someone who works with Perchance.

That's very rude of you.

It seems a lot of people are downvoting this post without even bothering to answer the question. I believe the older AI image generation model used some variation of the Stable Diffusion model. The new model is based on FLUX.1 schnell (according to the dev themselves).

What about getting the AI to only speak in a specific way in dialog?

For example, I have a character that is from Fire Emblem Awakening, his name is Gregor. and a key part of his personality is that he speaks from the third person and speaks with a Russian accent.

The thing is, I only want the AI to use the "third person speaking" for speech only, I even specified it in the AI reminder note, but for some reason, the character ALSO speaks in the third person in actions as well, even though I only want him to use the third person speaking in speech only.

This is an example (taken directly from one of my chats with the character involved)

AI: Gregor laughs big belly laugh at fisher-man clutching fish like baby! "Ah, but friend Rexy—Gregor thinks fish is smelling... strange? Like old calcaires in cave, nyet? Maybe Robin-man right about study. Gregor once saw fish with three eyes near Plegia border—tasted like regret and made Gregor's cousin glow purple! Not good eating!"

Is there a way to get the AI to only speak in the third person and with the Russian accent in dialog only and not actions?

I think this speaks for itself.

I mean, if you're going to volunteer to create this all yourself, by all means, do so. I sincerely doubt that the dev will actually do this, and if you do end up doing this all yourself, you're going to hit a few roadblocks:

• Recruiting moderators: You'll have to recruit users to be moderators for this forum and make sure you can trust them, make sure they won't power trip, unfairly block or ban users, or just be a dickhead in general.
• Running the forum in general: You're going to have to be spending a lot of hours on the actual forum itself to moderate it yourself when moderators aren't online, actively banning people who are causing trouble.
• Promoting: You're going to have to find some sort of way to promote the forum. You're obviously not going to pull everyone away from Lemmy, but that's fine. You don't want anti-AI people so that's kind of the point.

While this might seem a little difficult, it's not entirely impossible. Many generators on Perchance are created by users like you anyways, not everything is made by John Perchance himself, he just writes the code. You're the ones who use the code to your own liking.

I'm not being a spoiled baby. You're lucky you don't have to go through the same bs I'm going through. If you're going to insult me, might as well report yourself because the rules literally say to be kind and friendly lol.

Something called ignoring.

The title of this post is so out of pocket bro💀

"Join" bro thinks Perchance is some sort of golf club or something.

Why are you breaking the rules of Perchance forum? Swearing is against the rules and I will have to report this comment alongside everyone else who swears here.

Hi! A bit of an update to my situation. I made the hard choice of erasing all my data since my characters and everything became basically unrecoverable. Fortunately, I did have most of my characters saved on a Google Doc and I recreated a few of the ones I already had on an incognito tab and I'm going to import them over and start fresh. Not how I wanted this to go, but I had no choice :(

You know you can edit that out, right? People like you make such a big deal about the AI doing stuff that you don't want it to do... but even the dev himself literally says that if the AI doesn't act the way that you want it to, you can edit the AI's messages. If you edit the messages enough, the AI will act how you want it to.

You can also make use of the custom writing instructions as well as the character reminder note.

Bottom line, don't overreact, just edit the messages. Not that hard.

I believe Chloe is an original character made by someone on Perchance, the same goes for the other example characters (except Ganyu, who is obviously from Genshin Impact, Nanami who is from Jujutsu Kaisen, and Death who I believe is based on the interpretation of Death from Puss in Boots). Likewise, I believe her photo was generated using Perchance itself, though the exact prompt on how to generate a photo exactly like Chloe is unknown.

Fun Fact: Chloe actually used to be called "Assistant" before her name was changed to Chloe. You could still see her name as "Assistant" if you had data from like 1-2 years ago.

Use a website to make the image you want transparent.

I have reason to believe that the AI defaults to certain names and tends to make use of specific first and last names when generating characters. You can easily see this in the AI Character Generator when using the "Persona/Chat" feature that basically turns your generated images into actual characters you can chat with in the ACC. This only really applies to the old model, the new model is actually more creative with names.

Mainly, when generating characters with the "anime" art style, it'll generate Japanese names for the character. Or if generating furry characters (e.g. an anthro wolf), it just defaults to generic names like Luka or Kaira or Kara or some variation or form.

But I assume you're not using this feature but instead using the "Unknown" character that automatically generates characters based on a prompt. One pattern I DID notice here was that most characters had the last name "Castellanos" which was weird but I didn't really care much for it.

I agree. The new model isn't very good. The main issues I have with it are that it tends to break immersion, saying stuff like "BREAKING_BAD_PATTERNS" or talking about "breaking bad patterns" for no given reason whatsoever and has no context to the roleplay at hand. They also tend to speak like this all the time:

John nods, "I think we've got it under control," he said "but watch your back just in case."

Even if the initial message has asterisk roleplay such as this, characters added to that thread will still talk like the above example, asterisks have to be incorporated a LOT into the initial message (at least from what I observed) in order for the new characters to RP like that.

The new model isn't that bad though, as it is able to portray fictional characters almost perfectly. I don't really use the AI Story Generator much, I only use the AI Character Chat and occasionally the image generator. I'm saying this based on what I've observed from there.

I dunno what you're putting in the prompt. Could you perhaps provide example prompts? For example, I put this as the prompt:

Adult girl, brown hair, blue eyes

And what do you know, it generates photos of adult girls with brown hair and blue eyes. No underage content whatsoever. I don't know why you're specifically getting underage content unless you're putting in specific stuff. It may be an issue on your end, because for me, it works just fine.

Because it literally says "You have no rights" so you're implying that no one has rights.