Download over 2K before updating the next version.
[if no one uses means my checkpoint is not welcomed, thus I don't need to waste money and time on checkpoint.]
Advanced Version is a V1.0+, There will be more contrast and detail.
Trained on Over 2G of pictures carefully selected by PAseer and manually labelled.
Pursuit of generalisability and soft switching between different styles.
If you don't know what to use as a base model, use this.
Description
FAQ
Comments (3)
Hi. I like the pictures in your gallery and so i give it a try. Even if there is no info about compatible/preffered samplers, useful/best cfg scale, steps, or any additional info at all....... I mean, your goal is about 2k downloads. So hows about telling people what they will get, and what your model makes so outstanding,and how they should use it. Now i have to find it out all alone. ;-) Best regards
Thanks for liking it. What's special about this checkpoint is [featureless], no matter how you use it, it should match as long as it's not the kind of setting that will fail on any model.
So it's flexible for prompts, and for LORA, so no matter if you use a traditional sampler, or LCM attach a LCM-lora, or Turbo attach a Turbo-lora, or attach a Lighting-lora, you can generate pictures without compromising the quality, without compromising the natural language recognition.
After all, It's very similar to SDXL-base [after all, it's based on that training], and you can think that as a quality-enhanced version of SDXL-base.
And as a trainer, I honestly don't see any advantages by myself, instead there are a lot of shortcomings and problems, shortcomings in semantic recognition, shortcomings in knowledge of specific concepts, problems in the quality of generalisation for generating pictures of things with niche topics or themes, and so on. Because of the low input data.
One of the reasons for posting this model is I hope you can help me to test it to find the limitation and give your feedback. It's important to me and allows me to know where I can improve.
@Aseer Thx for your answer :-) Amazing news. You should write that in the discription of the model (flexibility,ect.,cos that sounds very good for people).
I am a businessman so thats why i am always complaining ;-)
(And thats why I always want to know what i will get :-) )