Posts

@truekry
About a month ago i personally did comment a bit on images that were posted here, but i only pointed out the positive stuff of what i liked on a picture. Mostly because i am unsure if the creator of said image would appreciate feedback on their images.

@truekry
I agree on missing feedback. Being told something as simple as “this is a cool idea” or “I like how this came out” would be nice.

@truekry
Hear here!

The worst thing about the silence is the missing feedback in my opinion. What did the people like about a picture, what did they dislike? Why did you get a down vote, why and up vote? It hard to guess. The only thing you can do for sure most people will like is plot to the face. I know, sex sells and all that, but I find it sad that we have this powerful tool that is AI, but if you want people to notice your art it better have plot in it. It also gives tantabus the vibe of a AI pony porn site and less AI pony art site.

I don’t mind the natural language prompting as much, since I’ve gotten used to Chroma, but the style clusters seem to both be necessary and a total mess and it’s pretty unreliable. I’d personally rather have a version of Chroma fine-tuned on ponies, or something similar.

Someone finetuning ponyv6 or something illustrious/noob based so that 90% of the pony character loras weren’t needed would be great as well.

(Another model that has some promise is NetaYume Lumina, but, IIRC, ponies didn’t come out very well on it when I tried it.)

Hopefully pony v8 being based on Qwen will end up better…

Well, from what I remember, there was going to be version 6.9 based on SDXL, but it didn’t happen. Using AuraFlow as a base seemed like the best option at the time, since there was no real alternative like Flux Schnell or Qwen then.

Putting issues with development and deployment aside, the current PonyV7 version seems like a fundamentally broken model hampered by the initial decisions and the lack of proper documentation. Focusing on the natural language prompting instead of tags made the model a chore to use, requiring a ridiculously long and flowery prompt to get any useful results. Using style clusters instead of artist tags looked like an interesting idea, but it turned out to be an utter mess, considering no documentation on these clusters and them clearly not working as intended.

The end result is a model that is slow, inconsistent, difficult to use and creates outputs that are SDXL-level at best and SD 1.5 level most of the time. I don’t see any reason to use it over the current Noob-based, or even PonyV6-based models, which are better in pretty much every single way.

I’m curious to know what this community thinks of ponyV7.

Personally, I think it’s a good model, but it should have used SDXL as a base.. Training loRAs is impossible (for me atleast, as someone who uses a AMD 16gb GPU on windows + has 32gb RAM), AuraFlow isn’t… the greatest base model… if they NEEDED to use a model that isn’t SDXL, then they could’ve used FLUX.. etc

In short, they probably should have used SDXL as the base model, but that’s my personal opinion.

What do y’all think? :o

Close up of mare parts, basic gooner nonsense.

@Bendy and Boney
An absolute classic!

Twilight Sparkle inquires about extra credit.

@NowIveAwoken
The flood of unedited gens is what got AI art removed from Derpi in the first place.

@Nocturn
Thank you.

We had a boo-boo. It’s been fixed. If you still see problems please let us know.

I wanted to see some horse ass.

all images are broken RIP

AI could provide better quality graphics than human art, but it alone doesn’t make generated images more valuable. Especially since AI can generate countless amount of such pictures. There are simply nothing unique, nothing memorable about them. And AI people doesn’t make things better generating dozens of basically same AI artworks.

A meme, a joke or a story incorporated in works of traditional artists is not a compensation for a less quality (it could be so but it’s not the main goal). It’s a thing that make artwork interesting, impressive. Good look on itself rarely make AI image impressive, especially among lots of other relatively good looking AI generated images.

@Felagund
Yeah, that is a good one. Didn’t fav because it isn’t THAT unique, nor is it a likely candidate for my D&D game since that is nonanthro Ponies (Capper and his like exist though).

Armored Princess Flurry Heart.