Hello everyone!
I want to speak my mind a bit and tell you what's what.
But first of all, I want to thank everyone who actively buys early access to my LoRAs! Thanks to you, they come into existence. And why? I'll explain a bit later.
I also don't want to leave out those who actively like my posts and write kind words. A big thank you to all of you too, from the bottom of my heart!!
Well then. I'll start pouring out my thoughts on the fact that lately I've been gathering a lot of datasets for training LoRAs and I'm a bit tired of it all. And I want to take a break from all of it – from training LoRAs and gathering datasets. On top of that, I'm also really tired of sorting and preparing datasets for training. I'm just tired, but on the other hand, no one is forcing me, and I'm very interested myself to see what comes out of all this, and if you guys like it too, then it makes it even more interesting for me to do.
Maybe I'll finally play some computer games. For now, I'll slowly release what's already ready and finish the LoRAs that I want to finish.
I've already accumulated 28 LoRAs, but I'm afraid that only about 12 of them will be on the site, and I myself will only use 2-3.
Now about early access.
I wouldn't use early access like many do and would just release LoRAs for free. But the problem is, I still can't set up the programs for training LoRAs on my PC. It's just a nightmare; I was dancing with a tambourine every time I tried to start the process, I tried many options, but it always comes down to either the LoRA not training properly, or errors popping up non-stop, or even worse, the computer stops responding, although that's very rare.
And that's exactly why I use the built-in function on Civitai for training LoRAs. That's why every donation and every early access purchase gives me the opportunity to train a new LoRA.
And a huge thanks to all these people!!!
Now I want to whine to you a bit.
I'll tell you from the very beginning.
2 years ago, when I started using AI generators, I posted all my pictures on Reddit, and to my surprise, after some time I already had 30k subscribers, I was infinitely happy about it. But at one point, Reddit put me in a shadow ban, even though I didn't spam or post controversial content. I also always asked the administration if I could post AI content, and when I got a positive answer, I quietly and calmly posted it and got my tiny piece of fame. But in the end, I was hidden behind a curtain. And I thought, okay, no big deal, and wrote to the Reddit admins thinking the system placed me there by accident. But the next day I was permanently banned for good.
It upset me a lot. And the most annoying thing is that I lost all contacts with people I had agreed to collaborate with, namely using AI to improve pictures or to help girls who were blurring their faces or covering them with a square. I was extremely surprised that they agreed to it and we were working together, but then everything was lost.
After six months, I created a new account and set it aside for an indefinite time. And just a week ago, I started slowly posting pictures where it's allowed (like 1-2 in AI groups), but today I was banned for life again. I don't understand what's wrong with Reddit, but I didn't break any rules, neither of the communities nor theirs.
Oh well. If that's how it is, then so be it.
In this chapter, I want to say that my subscribers are the best!
Surprisingly, I get quite a lot of messages saying that I've created something they really wanted, that they really like my LoRAs, or just that everything is super. Thank you! Even though I don't take things I read on the internet too close to heart, I am really surprised that someone likes all of this. And that can't help but make me happy.
And to those who write me negative messages, you are also the best. Without you, everything would be too good and I wouldn't see my own mistakes head-on. There aren't many of you, but you exist, and that's a good thing.
Now, regarding Pony models, Flux, Chroma, etc.
It just so happened that the SDXL niche was quite empty, and I decided to dive into it headfirst. I really like the MoP and Big Love models. Their creators, @SubtleShader and @GBRX, are impeccable. Thank you to you too!
Well, anyway... all the Pony models are already out there. I don't like Flux because it takes too long to generate anything on my setup, and Chroma looks interesting but is very raw, and it's unclear if it will be developed further. And I don't know the other models.
So, I'm not going to make anything for these models.
Fun fact:
This text was translated using AI!
Lately, I've been completely immersed in the world of AI and have started using it very actively in my life.
In conclusion, I want to say that I'm tired of writing and pulling words out of my head. I didn't even write this much back in university.
And why have you read all the way to this point? I would have stopped at the table of contents.
Thanks again, everyone. You are the best to me.