Update README.md

README.md

@@ -9,13 +9,26 @@ It can be used by modifying the `instance_prompt(s)`: **sololeveling**
 You can also train your own concepts and upload them to the library by using [the fast-DremaBooth.ipynb by TheLastBen](https://colab.research.google.com/github/TheLastBen/fast-stable-diffusion/blob/main/fast-DreamBooth.ipynb).
 And you can run your new concept via `diffusers`: [Colab Notebook for Inference](https://colab.research.google.com/github/huggingface/notebooks/blob/main/diffusers/sd_dreambooth_inference.ipynb), [Spaces with the Public Concepts loaded](https://huggingface.co/spaces/sd-dreambooth-library/stable-diffusion-dreambooth-concepts)
 
-This model was trained using 71 training images, 14200 total training steps, model saved every 3550 steps (25%) and text encoder was trained up to 35%. Made using Stable Diffusion v1.5 as the base model.
-
 This is my first model, criticism and advice is welcome. Discord: "Classacre#1028"
 This model is inspired by @ogkalu and his comic-diffusion model (https://huggingface.co/ogkalu/Comic-Diffusion). I think its pretty cool and you should check it out.
 
 I've made this model out of admiration towards Jang-Sung Rak (DUBU) who recently passed away. This model is not perfect, and will never be perfect as the original artists art is irreplaceable.
 
+### Version 2.0 ###
+This is a massive improvement from the first version. I've split the model into two different models, one for non action generations (SoloLevellingCalm.ckpt) and one for action generations (SoloLevellingAction.ckpt). I plan on merging the two into one model in the future once I understand how to do captions. The calm (SoloLevellingCalm.ckpt) version of the model is great for general generation using most prompts, it was trained using non action images taken from the solo leveling manhwa.
+
+**Important Prompt Additions:**
+Add these prompts to make the generations look remotely like the solo levelling art style and to maintain consistency.
+Positive prompts: anime, manhwa, beautiful, 8k
+Negative prompts: chat bubble, chat bubbles, ugly
+
+This model suffers from chat bubbles and added VFX words in its generations, it can often be mitigated by inputting the negative prompts in the Important prompt additions but it is not perfect. 
+
+Sampler and CFG settings are identical to Version 1.0. 
+
+### Version 1.0 ###
+This model was trained using 71 training images, 14200 total training steps, model saved every 3550 steps (25%) and text encoder was trained up to 35%. Made using Stable Diffusion v1.5 as the base model.
+
 The final model struggles to do calm / peaceful environments as it was trained on mainly cinematic action scenes - this leads to style bleeding where the ai creates action sequences from seemingly calm and peaceful prompts. Earlier models don't seem to have this problem albeit they are not as sharp and do not reproduce the style as accurately. Negative prompts seem to lessen the effects of action sequences in the final model, however they are not as natural as older models. Another thing to mention is that the model struggles at drawing eyes in action sequences, you may be able to play with the prompt to get eyes to show up though. A comparison between the different model versions can be seen below:
 
 Sampler used: DDIM