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!
Prompting:
***If you end up with a totally fried output image (extremely deformed or just noise), try adding or removing a few tokens to/from your prompt until you get a normal output. This is a very rare occurrence, but I did experience it once or twice during my testing. It may be related to very occasional bad luck when using weighted prompts.
Positive Prompt:
No special positive prompt is totally necessary, but to maximize realism, I recommend adding something like the below to the beginning of your prompt. You don't have to use this prompt string exactly, just something in the same ballpark that pushes the model towards a more realistic, photographic style.
RAW photo, DSLR, 35mm film grain, photorealistic, hyperrealistic, candid photograph, shot on Sony A7III, 85mm f/1.8, shallow depth of field, Negative Prompt:
Again, no special negative prompting is really necessary, but you can use something like the below to push the model further away from non-realistic art styles. Like above, you don't need to use this exact prompting, just something in the same spirit.
anime, illustration, illustrated, cartoon, animated, animationAt a Glance:
Style: Realistic
Works with LoRAs?: Technically yes, but if it actually looks good may be hit or miss
VAE: sdxl_vae is baked (but you can override if desired)
Suggested parameters: DPM++ 3M SDE, Karras, 40 steps, 832x1216/1216x832, CFG 3 (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 (probably more necessary here than with an anime base model)
Description:
I've been trying to make a realistic Illustrious merge for quite a while now, but I could never get results that actually looked good. I finally stumbled onto a concoction that was very promising, but some of the models in the merges didn't play nice with each other and caused some prompts to occasionally get fried and output garbage. So I refined from there, and while I wasn't able to totally eliminate that random model failure, I got it to the point that it seems to be rare enough that I'm comfortable sharing this model family now. A brief description of each model is below.
General: A jack-of-all-trades. If you know you specifically want an Asian or Western appearance, then it might be better to use one of those respective models, but if you don't really care and just want to mess around with something flexible and generally capable, then go with this one!
Western: This one aims to create a general Western (non-Asian) appearance. You may be able to prompt for specific nationalities/ethnicities, but it might be hit and miss.
Asian: This one aims to create more of a general Asian appearance. You may also be able to prompt for specific Asian nationalities/ethnicities, but it may not always work.
Unstable Versions: These are relics from early on in the merging process. They are a bit more likely to fry your prompts (still relatively rare though), but when they work properly, they can sometimes produce even better results aesthetically than the final versions. No promises with these though, just treat them as experimental and play around with them! There's no unstable version of the Asian model, as that one worked pretty well right off the bat.
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!