ItN's Illustrious Merge

I am admittedly kind of a noob at this, but if you enjoy my humble work, any donations are greatly appreciated: https://ko-fi.com/insult_to_ninjary

Your generosity helps me stay motivated to keep improving and to keep trying new things!

At a Glance:

• Style: Anime/Illustrated

• Works with LoRAs?: "Base" versions - should be fine, "Plus" versions - mostly yes but maybe not certain style LoRAs

• VAE: sdxl_vae is baked (but you can override if desired)

• Suggested parameters: Euler a, Automatic, 30 steps, 832x1216/1216x832, CFG 5 (but feel free to experiment!)

• Hires. fix: 1.5x, use a quality upscaler (like 4x-UltraSharpV2), denoising strength 0.5 - 0.3

• Extra: use ADetailer to fix faces and/or hands if necessary

Description:

I know I said that I was done making merges, but since I just started getting into Illustrious, I decided to try my hand at a few Illustrious merges. But don't expect a massive amount of them this time around. Unlike with SD1.5, where I made a ton of merges with a bunch of different aesthetics, my goal for Illustrious is just to have one core neutral/general model, with a small number of variations of it for different preferences or use cases. For the foreseeable future, I have 4 variants that I'll detail below.

Default: [Coming soon!] This is an extended merge of the "Cute" version, which somewhat cuts down on the surface-level cuteness and makes it a bit more appropriate to use as a general purpose anime base model. However, you can still coax out cuter images with appropriate prompting. So it has some nice flexibility. The model looks great when used alone with no LoRAs (if a little generic), but can also be used as a style-neutral base model for (hopefully) pretty much any LoRA you want to throw at it. Because it's flexible enough to be used both alone or with LoRAs, and usually looks great in a variety of situations, I'm going to set this one as the main model that shows up first on the model page. However, if you're not interested in using LoRAs at all, I recommend "Default Plus," and if you're specifically looking for something that leans more cute right out of the box, then check out the two "Cute" versions.

Default Plus: [Coming soon!] This is an extended merge of the "Default" version, including a few extra LoRAs in the mix (I think). Compared to the "Default" version, this one seems to produce a bit more detail and depth, and I think it looks better than the "Default" version when used alone with no LoRAs. However, it may be a bit less flexible. And while you can generally also use LoRAs with this version, please note that certain style LoRAs, especially ones that have a drastic effect (such as pixel art LoRAs, for example), might not work that well. I think it's a bit more accepting of LoRAs now than "Cute Plus" is, but probably not perfect still. Character LoRAs, and most other types, should mostly be ok tho.

Cute: [Coming soon!] The goal of this one was to be a general purpose anime model, but it came out leaning a bit more cutesy than I intended. Aside from that, the model still looks really good (tho maybe a bit generic), and otherwise has a pretty neutral style that should work well as a base for LoRA use. You may be able to prompt out some of the cuteness if you're a better prompter than me, but as it stands, I think the model came out good enough to share, even if it didn't quite turn out as I expected.

Cute Plus: This is an extended merge of the "Cute" version, including a few extra LoRAs in the mix (I think). Compared to the "Cute" version, this one seems to produce a bit more detail and depth, and I think it looks better than the "Cute" version when used alone with no LoRAs. However, it seems to be a bit less flexible. And while you can generally also use LoRAs with this version, please note that certain style LoRAs, especially ones that have a drastic effect (such as pixel art LoRAs, for example), might not work that well. Character LoRAs, and most other types, should mostly be ok tho.

Troubleshooting:

If you're having persistent problems with anatomy (hands, fingers, etc.), fine detail, composition, and/or other jank, just try your best with all the usual fixes - different prompting (including negative prompt), use a different VAE, use adetailer, inpaint, make sure to try the above suggested parameters, etc.

Outro:

If you make anything awesome, please consider posting them to the gallery! That goes for all of my models! I really like seeing what kind of stuff you all are able to make!

If you have any questions, feel free to ask! I will try my best to help! Also, if you have any ideas for future models, please don't hesitate to share them with me!