Moisturize things with water, and hold colors in the heart.
[ENG Instructions below]
Please note that the tendency of each version is different, if you like the most transparent watercolor effect, please choose v1, but it is suitable for relatively few models.
v4 is currently the most recommended version.
Other versions have their own evolutionary direction, if you are interested, welcome to try more!
Please note that each version has a different tendency.
If you like the most transparent watercolor effect, please choose the original v1, but it fits relatively few models.
v4 is the most recommended version right now.
Other versions have their own evolutionary direction, so feel free to try more if you're interested!
This is a watercolor-style LoRA, which is still being improved, and all kinds of feedback are welcome!
This is a watercolor style LoRA, still in the process of improvement, welcome all kinds of feedbacks!
The v4 version has greatly improved the ability to adapt to multiple models, so I won’t go into details, please refer to the sample picture and you will understand it immediately. Similarly, it can work with a large number of other excellent lora, just pay attention to the combination weight.
Recommended settings: Weight=0.8-1, CFG=3-6
The v4 version is a great improvement in the ability to adapt multiple models, so without further ado, please refer to the sample image and you will understand immediately. Likewise, it can work with a large number of other lora, just be careful with the combination weights.
Recommended settings: weight=0.8-1, CFG=3-6
General model recommended configuration: CFG, weight 0.65-0.75, CFG is completely linked to the color shade, but please note that depending on the model, some overfitting will occur around 7. But some three-dimensional models will have a more retro heavy watercolor performance on 7.
When using with other LoRA, please pay attention to avoid weighting and too high. Generally speaking, the total weighting value should not exceed 1.2. If you need to show more watercolor styles, you need to keep the weight of Qincai as the highest one.
Example: The picture below shows the effect of using Moxin and Shuke Zouma with a total of 3 LORAs, and the weight ratio is 0.5+0.35+0.35
Instructions for v3
Generic model recommended configuration: CFG, weight 0.65-0.75, CFG is fully linked to color intensity, but please note that depending on the model, some overfitting will occur at around 7.But there are some realistic models will have a more vintage heavy watercolor performance on top of 7.
When using with other LoRA, please pay attention to avoid weighting and too high, generally speaking, the total weighted value is not more than 1.2 is more appropriate, if you need more watercolor style to appear, you need to keep the weight of the QIN color as the highest one.
Example: The image upstairs shows the effect of using Coloerwater, Moxin and Shukezouma total of 3 LORA, the weighting ratio is 0.5 + 0.35 + 0.35
The current Qincai has undergone various enhancements, and has a certain realistic ability!
The recommended configuration of the two-dimensional class model: CFG, weight 0.85-1
Three-dimensional class model recommended configuration: CFG, weight 0.5-0.7
At present, the style of the character's face still retains the original style of the model to the greatest extent. The v3 version will try to improve this part and pursue a realistic style. At that time, the two-dimensional and three-dimensional versions will be separated for everyone to try.
Thank you for your good comments!
Now Colorwater has undergone various enhancements and has a try on realistic style!
Anime-like models: CFG, weight 0.85-1
Realistic-like models: CFG, weight 0.5-0.7
At present, the style of the character's face is still the maximum to retain the original style of the model, v3 version will try to improve this part, the pursuit of realistic style. If things go well, there will be two diff versions by the time they are completed for everyone to try.
Thank you all for your kind comments!
It is recommended to use a weight of 0.85-1, if you use it with other lora, please judge by yourself.
Due to some reasons, the data set deliberately excludes face-related data, so the performance of most of the main models on the face of the characters is still its original characteristics, please adjust it according to your needs.
The style of watercolor will change according to the value of CFG, it is recommended to experience different variations from 2~6. For higher values, please pay attention to whether it can match your model. In general, the high CFG is positively related to the intensity of the watercolor.
The weights are recommended to use 0.85-1, please use your own judgment if matching with other lora.
For some reasons, the dataset deliberately excludes face-related data, so most of the main models in the character face are still their own original features, please just adjust yourself as needed.
Visit Official Website