IllustMixClarity is a semi-realistic merged model with an integrated VAE, built upon the Illustrious-series of models.
To maximize the model's performance, it is recommended that prompts be described in as much detail as possible. (It might be a little temperamental at times.)
To emphasize a realistic depiction, we recommend including the keyword realistic in your prompt. Conversely, if you intend to achieve an anime-style expression, please omit this keyword and adjust accordingly.
This repository is the official release version of the model previously published as "Noginowa/semireal_merge_experiment". The model has been renamed for the formal release, and the filename has been changed from semireal_merge_experiment_[version].safetensors to IllustMixClarity_[version].safetensors.
For supplementary information: The current version, IllustMixClarity_v1.safetensors, has the identical content as the former final adjustment version, semireal_merge_experiment_v25112601.safetensors. It was formally released after being renamed.
❗ご利用開始前に必ずお読みください：最適な結果を得るため、以下の「おすすめ設定」と「おすすめプロンプト」を必ず確認してください。 ❗MUST READ Before Use: To achieve optimal results, please ensure you check the "Recommended Settings" and "Recommended Prompts" sections below.
We do not actively recommend the use of Hires. fix with this model.
Please use it only when necessary, such as when high resolution is required (e.g., when characters are rendered small).
✨ おすすめ設定 / Recommended Settings
設定/Parameters
内容/Values
CLIP skip
2
CFG scale(Euler a)
3-7
CFG scale(Euler a CFG++)
1.5
Steps
20-40(おすすめ/Recommended 28)
Sampler
Euler a, Euler a CFG++
🖼️ サンプル画像 / Sample Images
サンプル画像はv1で生成しています。
Sample images were generated with v1.
プロンプトにRealisticを含めた例 / Example with "Realistic" in the prompt:
プロンプトにRealisticを含めない例 / Example without "Realistic" in the prompt:
💬 おすすめプロンプト / Recommended Prompts
Positive:
Realistic, Masterpiece, Best quality, Amazing quality, Very aesthetic, Absurdres, Newest
お好みに合わせてこれらを組み合わせて使用すると、見た目や質感がよりリアルになります。(すべてを含める必要はありません)
You can combine these options as desired to achieve a more realistic appearance and texture. (You don't need to include all of them)
Negative:
Nsfw, Lowres, Bad quality, Worst quality, Poorly drawn, Flat color, Sketch, Jpeg artifacts,
Bad anatomy, Ugly, Bad eyes, Bright pupils, White pupils, Bad hands,
Blurry, Furry, Watermark
📝 サンプルプロンプト / Example Prompts
Positive:
Depth of field, Portrait, Face focus, Dynamic angle,
1girl, Solo, Light smile, Closed mouth, Light blush,
Long hair, Hair between eyes,
White shirt, Collared shirt, Bowtie,
Outdoors, Fashionable street, Cafe, Crowd of people,
Realistic, Masterpiece, Best quality, Amazing quality, Very aesthetic, Absurdres, Newest
Negative:
Lowres, Bad quality, Worst quality, Poorly drawn, Flat color, Sketch, Jpeg artifacts,
Bad anatomy, Ugly, Bad eyes, Bright pupils, White pupils, Bad hands,
Simple background, Transparent background, Blurry, Furry, Watermark
🛠️ マージ情報 / Model Merging Details
本モデルの詳細なマージ比率や使用したベースモデルの情報については、以下の専用ファイルをご参照ください。
For detailed merging ratios and information on the base models used, please refer to the dedicated file below. ➡️ 詳細なマージ情報はこちら (IllustMixClarity_v1.md)
