Hassaku aims to be a anime model with a bright and distinct anime style.
You can run Hassaku XL and use its API on SinkIn: https://sinkin.ai/m/JWknjgr
My Discord for everything related to anime models and art.
If you'd like to support my work, you can do so at: SubscribeStar, every bit helps and is truly appreciated!
____________________________________________________________
Supporters:
Thanks to my supporters tamashiicolle, pttcot, SETI and Kodokuna
____________________________________________________________
Using the model:
The model is trained using images with minimal disruptive elements such as floating text, logos, speech bubbles, and signatures. If any of these elements are present in an image, please include the prompt "signature" as a negative prompt
Metadata and franchise tags are excluded—please do not use them. Tags such as "highres" and franchise-related tags like "re:zero kara hajimeru isekai seikatsu" are not used in training
A CFG of 3 to 7 is recommended
Look at the example images, to see how the model should be used
Model is trained to be not complicated in use, tag order:
First are the number of person, then character names, rest.
Example: 1girl, rem \(re:zero\), standing, masterpiece, upper bodyNoobAIs Loras working for the most part better in compare to Illustrious loras
Here are some resolution options for SDXL:
1536 x 640
1344 x 768
1216 x 832
1152 x 896
1024 x 1024
896 x 1152
832 x 1216 (most recommended)
768 x 1344
640 x 1536
______________________________________________________
Version and License info:
Below V1 merges ANIMAGINE XL 3.0
Version V1 use Illustrious-XL & WAI-NSFW-illustrious-SDXL with additional training
V2 is trained on its own and don't include any extra merge. Base was V1, it is trained to include newer or missing characters. It was also used for further training tests
V3 is a merge of Illustrious-XL and V2.2, to fix some issues and was also trained to include newer or missing characters
All Models using the Fair AI Public License 1.0-SD license.
Description
The pure trained model output. Completly unaltered, as Nature intended.
This model does not have any style merge stabilisation and is intended as base for further training, merges, using Loras, etc. for anyone, that are interested in that.
Be aware that it can have every kind of problem, a base model can have.
But it have a little specialty to it, the style got natively trained into the model, instead of merging it into the model per loras. It enhance the default output of the pure model.
If you compare it with 3.1, you will notice some similar looking images, because of the low style merge of 3.1 itself. In some cases, this pure model may produce better looking images, but it can also create significantly worse images. Results will vary greatly.
It have one trained epoch more (400k images -> not a big differece) in, in compare to 3.1.
With slitly changed training:
Trained now on LION, with Stochastic Rounding, SMMF, C-Optim and higher Alpha1. Before with adafactor Stochastic Rounding and Fused Back Pass.