Support☕ https://ko-fi.com/sfa837348
hugging face https://huggingface.co/gsdf/CounterfeitXL-V2.0
Negative Embedding https://civarchive.com/models/118418/negativexl
Official hosting for online AI image generator.
A small update.
V2.5 is a revised version of V2.0.
This version has been retrained for style tags.
Tips
By inserting 'flat color, anime style' into negative prompts, you can achieve results with higher detail.
CounterfeitXL-V2.0 has been primarily trained on the latest illustrations but encounters a recurring issue with the appearance of halos. In such cases, please include 'halo' in negative prompts.
I have incorporated style tags into some of the training images. By describing these tags in the prompt, various effects can be achieved. Additionally, inserting style tags into negative prompts or combining them allows for the generation of images that align with your imagined.
Style Tags
masterpiece: Beautifully and intricately drawn illustrations.
cinematic style: Stunning illustrations with detailed backgrounds and lighting.
aesthetic style: Illustrations featuring beautiful color schemes.
cute style: Cute.
flat color: Illustrations with strong lines and a minimal color palette.
anime style: Anime.
comic style: Illustrations characterized by bold color choices and abstract representations.
Description
CounterfeitXL-V2.0 has been primarily trained on the latest illustrations but encounters a recurring issue with the appearance of halos. In such cases, please include 'halo' in negative prompts.
I have incorporated style tags into some of the training images. By describing these tags in the prompt, various effects can be achieved. Additionally, inserting style tags into negative prompts or combining them allows for the generation of images that align with your imagined.
Style Tags
masterpiece: Beautifully and intricately drawn illustrations.
cinematic style: Stunning illustrations with detailed backgrounds and lighting.
aesthetic style: Illustrations featuring beautiful color schemes.
cute style: Cute.
flat color: Illustrations with strong lines and a minimal color palette.
anime style: Anime.
comic style: Illustrations characterized by bold color choices and abstract representations.
FAQ
Comments (9)
I know that you have an "official online generator", but I'll ask any way 🙏.
Can you please enable the model to allow it to be used for civitai's online image generation? Thank you for your consideration.
Is it just me or does CounterfeitXL 2.0 have problems with very short prompts?
Any short prompt attempts, especially <6 tokens, often come out extremely faded as if I'd forgotten VAE.
This version seems quite volatile.
2.0 maybe more “Fits the description”，but the“The quality of the results piece” is worse than before
Sentence comprehension is better than its predecessor, v1.0.
Please do LCM+Turbo model
you can still use lora lcm+turbo, and it works as well
@mskiller51 which one?
@infrezz721 https://civitai.com/models/216190/lcmandturbomix-lora-only-12mb-8-step-sampling-effect-is-superior-to-using-lcm-or-turbo-alone this one for example
@mskiller51 can't they make their own Turbo/Lightning/LCM LoRA set for the same model? It might yield different results compared to the original SDXL Lightning LoRAs