docs: update README

.gitattributes

@@ -39,3 +39,4 @@ images/AnimaMixColorXL_v3_cover.jpeg filter=lfs diff=lfs merge=lfs -text
 images/AnimaMixColorXL_v3_v2_plot.jpeg filter=lfs diff=lfs merge=lfs -text
 images/SampleImage_AnimaMixColorXL_v3_1.jpeg filter=lfs diff=lfs merge=lfs -text
 images/SampleImage_AnimaMixColorXL_v3_2.jpeg filter=lfs diff=lfs merge=lfs -text
+images/AnimaMixColorXL_v2.1_v2_plot.jpeg filter=lfs diff=lfs merge=lfs -text

README.md

@@ -1,70 +1,74 @@
----
-license: other
-license_name: faipl-1.0-sd
-license_link: https://freedevproject.org/faipl-1.0-sd/
-language:
-- en
-pipeline_tag: text-to-image
-tags:
-- stable-diffusion
-- stable-diffusion-diffusers
-- text-to-image
-- merge
-inference: false
-library_name: diffusers
-base_model:
-- cagliostrolab/animagine-xl-4.0
-- cagliostrolab/animagine-xl-3.1
-- Noginowa/AnimaMixColorXL
----
-
-![Cover](https://huggingface.co/Noginowa/AnimaMixColorXL/resolve/main/images/AnimaMixColorXL_v3_cover.jpeg)
-
-Animagine XLベースのVAE内蔵マージモデルです。必ずおすすめ設定、おすすめプロンプトを確認してください。<br>
-This is a merged model based on Animagine XL with a built-in VAE. Be sure to check the recommended settings and recommended prompts.<br>
-<br>
-
-# ライセンス / License
-[Fair AI Public License 1.0-SD](https://freedevproject.org/faipl-1.0-sd/)<br>
-<br>
-
-# モデルについて / About the Model
-現在の最新モデルはAnimaMixColorXL v3です。<br>
-The current latest model is AnimaMixColorXL v3.<br>
-<img src="https://huggingface.co/Noginowa/AnimaMixColorXL/resolve/main/images/AnimaMixColorXL_v3_v2_plot.jpeg" width="500px"><br>
-
-# おすすめ設定 / Recommended Settings
-CFG Scale: 4-7(おすすめ:5)<br>
-Sampling Steps: 25-28(おすすめ:28)<br>
-Sampler: Euler-a<br>
-<br>
-
-# おすすめプロンプト / Recommended Prompt
-Positive:
-```
-Masterpiece, Best quality, Amazing quality, Very aesthetic, Absurdres
-```
-Negative:
-```
-Nsfw, Lowres, Bad quality, Worst quality, Low score, Bad score, Average score, Poorly drawn, Flat illustration, Flat color, Anime style, Sketch, Jpeg artifacts,
-Bad anatomy, Ugly, Bad eyes, Bright pupils, White pupils, Bad hands, Bad fingers, Extra fingers, Simple background, Transparent background, 3D, Blurry, Furry,
-Watermark
-```
-<br>
-
-# 詳しい使い方 / Detailed usage
-v3は[Animagine XL 4.0](https://huggingface.co/cagliostrolab/animagine-xl-4.0)、v2以前は[Animagine XL 3.1](https://huggingface.co/cagliostrolab/animagine-xl-3.1)と互換性があります。詳しくは各Animagine XLページのUsage Guidelinesを参照して下さい。<br>
-v3 is compatible with [Animagine XL 4.0](https://huggingface.co/cagliostrolab/animagine-xl-4.0), and v2 and earlier versions are compatible with [Animagine XL 3.1](https://huggingface.co/cagliostrolab/animagine-xl-3.1). For more details, please refer to the "Usage Guidelines" section on the respective Animagine XL pages.<br>
-<br>
-
-# サンプル画像 / Sample Images
-サンプル画像はAnimaMixColorXL v3で生成し、Hires. fixをしています。<br>
-Sample images were generated with AnimaMixColorXL v3 and Hires. fix.<br>
-<img src="https://huggingface.co/Noginowa/AnimaMixColorXL/resolve/main/images/SampleImage_AnimaMixColorXL_v3_1.jpeg" width="500px"><br>
-<img src="https://huggingface.co/Noginowa/AnimaMixColorXL/resolve/main/images/SampleImage_AnimaMixColorXL_v3_2.jpeg" width="500px"><br>
-
-# 作者 / Creator
-Civitai: [Noginowa](https://civitai.com/user/Noginowa)<br>
-Bluesky: [のぎのわ](https://bsky.app/profile/noginowa-ailab.bsky.social)<br>
-<br>
+---
+license: other
+license_name: faipl-1.0-sd
+license_link: https://freedevproject.org/faipl-1.0-sd/
+language:
+- en
+pipeline_tag: text-to-image
+tags:
+- stable-diffusion
+- stable-diffusion-diffusers
+- text-to-image
+- merge
+inference: false
+library_name: diffusers
+base_model:
+- cagliostrolab/animagine-xl-4.0
+- cagliostrolab/animagine-xl-3.1
+- Noginowa/AnimaMixColorXL
+---
+![Cover](images/AnimaMixColorXL_v3_cover.jpeg)
+
+Animagine XLベースのVAE内蔵マージモデルです。必ずおすすめ設定、おすすめプロンプトを確認してください。<br>
+This is a merged model based on Animagine XL with a built-in VAE. Be sure to check the recommended settings and recommended prompts.<br>
+<br>
+
+# ライセンス / License
+[Fair AI Public License 1.0-SD](https://freedevproject.org/faipl-1.0-sd/)<br>
+<br>
+
+# モデルについて / About the Model
+現在の最新モデルはAnimaMixColorXL v3です。  
+The current latest model is AnimaMixColorXL v3.  
+
+v2.1は、v2のレシピをベースに怪しい部分を修正し、sd-mechaでマージし直しました。  
+v2.1 was re-merged with sd-mecha after correcting problematic areas found in the v2 recipe.  
+
+<img src="images/AnimaMixColorXL_v3_v2_plot.jpeg" width="500px"><br>
+<img src="images/AnimaMixColorXL_v2.1_v2_plot.jpeg" width="500px"><br>
+
+# おすすめ設定 / Recommended Settings
+CFG Scale: 4-7(おすすめ:5)<br>
+Sampling Steps: 25-28(おすすめ:28)<br>
+Sampler: Euler-a<br>
+<br>
+
+# おすすめプロンプト / Recommended Prompt
+Positive:
+```
+Masterpiece, Best quality, Amazing quality, Very aesthetic, Absurdres
+```
+Negative:
+```
+Nsfw, Lowres, Bad quality, Worst quality, Low score, Bad score, Average score, Poorly drawn, Flat illustration, Flat color, Anime style, Sketch, Jpeg artifacts,
+Bad anatomy, Ugly, Bad eyes, Bright pupils, White pupils, Bad hands, Bad fingers, Extra fingers, Simple background, Transparent background, 3D, Blurry, Furry,
+Watermark
+```
+<br>
+
+# 詳しい使い方 / Detailed usage
+v3は[Animagine XL 4.0](https://huggingface.co/cagliostrolab/animagine-xl-4.0)、v2以前は[Animagine XL 3.1](https://huggingface.co/cagliostrolab/animagine-xl-3.1)と互換性があります。詳しくは各Animagine XLページのUsage Guidelinesを参照して下さい。<br>
+v3 is compatible with [Animagine XL 4.0](https://huggingface.co/cagliostrolab/animagine-xl-4.0), and v2 and earlier versions are compatible with [Animagine XL 3.1](https://huggingface.co/cagliostrolab/animagine-xl-3.1). For more details, please refer to the "Usage Guidelines" section on the respective Animagine XL pages.<br>
+<br>
+
+# サンプル画像 / Sample Images
+サンプル画像はAnimaMixColorXL v3で生成し、Hires. fixをしています。<br>
+Sample images were generated with AnimaMixColorXL v3 and Hires. fix.<br>
+<img src="https://huggingface.co/Noginowa/AnimaMixColorXL/resolve/main/images/SampleImage_AnimaMixColorXL_v3_1.jpeg" width="500px"><br>
+<img src="https://huggingface.co/Noginowa/AnimaMixColorXL/resolve/main/images/SampleImage_AnimaMixColorXL_v3_2.jpeg" width="500px"><br>
+
+# 作者 / Creator
+Civitai: [Noginowa](https://civitai.com/user/Noginowa)<br>
+Bluesky: [のぎのわ](https://bsky.app/profile/noginowa-ailab.bsky.social)<br>
+<br>
 The English translation of this workflow description was created with the help of Gemini.