[Megacloud Daily Upload Post] December 13th & Access Member Previews (Patreon)

Megacloud Link: (Accessing this site as an Access Member will be denied due to Sign In with Patreon lock for only cloud and up members)

https://rule34diffusion.com/rule34-diffusion/

Since I'm running out of characters to make for Main reQuest, that means the 2nd place choice as voted on by members will start to be made!

Hazbin Hotel Girls will start coming out for all Tiers! I will be redoing the character models of some of them as well and have the dataset ready to be sorted tomorrow!

[See at the bottom for post Talk]

==== [Anime Days Content] ====

[Brainstorm Form] Komachi Hikigaya Sexy Poses [Gen 2]

[Brainstorm Form] Komachi Hikigaya Full Variety [Gen 2]

Tohru V2 Sexy Poses [Gen 2]

Tohru V2 Full Variety [Gen 2]

Sailor Uranus Sexy Poses [Gen 2]

Sailor Uranus Full Variety [Gen 2]

Sailor Moon Sexy Poses [Gen 2]

Sailor Moon Full Variety [Gen 2]

[Video Game Content] (Last Images, I almost forgot to add them)

Ultimate Freyna (Other Form) Full Variety [Gen 2]

This one was a bit tough but not too bad, I just had to FORCE the helmet on her hahaha

==== [Character Commissions] ====

[Cloud Plus Tier Commission]

[Character Request] Aegis (Persona) Finished [Gen 2]

[Cloud Plus Tier Commission]

[Character Request] Secre Swallowtail [Gen 2]

[Character Request] Shutara Senjumaru Unfinished [Gen 2]

[Post Talk]

This is going to be about multi-character works.

I've been bothered by the quality of some of the works I've released, primarily because I've seen other creators produce much higher-quality multi-character content.

However, some of them are significantly larger and can likely afford a cloud GPU or simply a better GPU locally. One of the major issues I'm facing is that I'd like to increase the image generation size in order to improve the quality of my results.

However, it would take much longer unless I drastically reduced the number of images.

I've been trying to figure out how to get a cloud GPU, but the majority of the ones set up for image generation are a newer or different version of the webui I use.

As a result, the extension I absolutely need is incompatible with those versions.

So, over the next few weeks, I'll be looking for a way to set up a cloud GPU specifically for multi-character content in order to improve quality while not reducing the number of images I create.

Also of note, is that some creators utilize the built-in character prompts and use a style model for whatever anime they are doing to pull off the more complex stuff.

Like 69, scissoring (like mine, but with fewer issues) and more physical interactions.

Using Character LoRa models, they must be separated so that they do not affect the other characters physical features, which is why some of the complex ones mentioned above are so difficult.
So by splitting the image for each model, it treats each side as if it were a separate image, with the exception of the background, which is not split.