Inuyasha | 犬夜叉 - Collection

This LoRA is intended to train some of the aesthetics and styles of the Inuyasha and Yashahime series by mangaka Rumiko Takahashi.

The characters of Inuyasha and style of Rumiko Takahashi are already well learned by the base models - so this has mostly been trained on a dataset including fan works for emphasis.

Other are LoRA in most of the previews, so please reference this post for singular effect of this LoRA: https://civitai.com/posts/11723453

Character Tags:

inuyasha

inuyasha \(character\)
higurashi kagome
kikyou \(inuyasha\)

kouga \(inuyasha\)
kagura \(inuyasha\)

miroku \(inuyasha\)
sango (inuyasha)

naraku (inuyasha)
rin (inuyasha)
sesshoumaru
sakasagami no yura

han'you no yashahime

moroha
higurashi towa
setsuna (inuyasha)

Recommended prompt structure:

Positive prompt:

artist:{{artist}}, inuyasha, takahashi rumiko \(style\),
{{tags}}
masterpiece, best quality, very awa, absurdres

Negative prompt:

(worst quality, low quality, sketch:1.1), error, bad ???????, bad hands, watermark, ugly, distorted, censored, lowres, abstract, signature, bkub

Description

Trained on NoobAI-XL (NAI-XL) V-Pred 1.0-Version

With the kohya_ss dev branch, v_paratemization ,zero_terminal_ssr enabled, and noise offet set to 0.

Previews are generated in Forge with DynamicThresholding (CFG-Fix) Integrated enabled, settings:

dynthres_enabled: True, dynthres_mimic_scale: 7, dynthres_threshold_percentile: 1, dynthres_mimic_mode: Half Cosine Down, dynthres_mimic_scale_min: 1, dynthres_cfg_mode: Half Cosine Down, dynthres_cfg_scale_min: 3, dynthres_sched_val: 1, dynthres_separate_feature_channels: enable, dynthres_scaling_startpoint: ZERO, dynthres_variability_measure: STD, dynthres_interpolate_phi: 1

A ComfyUI workflow reference with DynamicThresholding, Upscaling, and FaceDetailer can be found here: https://civitai.com/posts/11457095

Inuyasha | 犬夜叉 - Collection

Description

Details

Files