Post your gens down below! I will ⚡ images I like! Have fun! <3
Positive: masterpiece, very aesthetic, best quality, 8k, Anime-art, amazing quality, very awa, absurdres, newest, high detail, cinematic lighting, (impressionism:1.5), high detail, depth of field, (blurred background), (dramatic lighting), atmospheric, dynamic angle, volumetric lighting
Negative: (worst quality:1.3), muscular woman, deformed, distorted, disfigured, bad anatomy, missing legs, extra legs, extra limbs, cloned face, low resolution, low quality, jpeg artifacts, (watermark, signature, text:1.5), error, ugly, bad proportions, bad face, unrealistic, mutated, cross-eye, fused eyes, poorly drawn face,obese, Chubby, curvy, over weight, overweight, sweat,
Description
FAQ
Comments (18)
I THINK I'LL LOSE MY MIND IN HYSTERIA
I hope machine girl is longer!
Only model I've found where LCM actually is pretty good, try it out guys I do lcm and beta_1_1 (using artsy)
love it, perfect for creepy mixed with cute
Which upscaler do you recommend for the 'artsy' checkpoint?
hey! if you like the grainy style of it i suggest using 001_classicalSR_DF2K_s64w8_SwinIR-M
to have things cleaner you can use anime 2b or something similar
@Shiiro0 Marvelous, thank you!
@Shiiro0 late to the party, but thanks for reccing this upscaler, i'm always on the lookout for ones that preserve texture
This model is the greatest.
Biggest thanks.
This is one of the best checkpoints! It understands many of the keywords and styles that often other checkpoints fail to.
Thank you for making it!
Please keep up the great work ^^
Figured I'd replicate my review here so it gets more viewership from those who might want to check it out. Such a fantastic model, I've been using this recently and trying new styles. It renders cartoon styles amazingly well, being malleable enough to easily adjust from more realistic styles, to deep blocky color shading. Works excessively well as both a refiner and base model, and plays well with all the LoRAs I've tried from Pony and Illustrious. Simply put, if you haven't tried this model, perhaps you're missing out.
This is my main checkpoints right now lol. Prompt coherence gone crazyyyy
The best anime checkpoint I've used so far. Thank you!
It is a massive travesty that the "artsy" checkpoint isn't more popular. I'm literally upset about it. I need more images with it! I wanna see what other people have made!!
I use it all the time and I was super disappointed to see my modest follower base on social media (8k~) didn't care for it compared to typical ""slop"" style stuff I typically make. This checkpoint easily has the least "AI Art" "look" I've ever seen.
I can't sing enough praises. Wonderful, wonderful checkpoint. Thank you so much. At least I can enjoy it myself.
PS- Since I'm an educator and will be teaching a class on SD soon, I will enjoy forcing my students to use this checkpoint! >:)
While it's indeed good, the style is too monotonous and not sensitive enough when it comes to using artist prompts
who is artist?
The context window of this model is 77 tokens, after which the value of the promt begins to be lost. Quality tags occupy 99% for industrial products, great