Good to see you! Try the new version!
(5/11/2005)
Today we newly released v7.0,
Sample images are generated at 720 x 1280,( Upscaler is R-ESRGAN 4x+ Anime6B 720x1280 to 1080x1920) step 25, Euler a/SGM Uniform. ADetailer and CD Tuner available, VAE is baked in. Basically not necessary,
サンプル画像は720x1280で生成されています（アップスケーラはR-ESRGAN 4x+ Anime6B 720x1280→1080x1920） step 25, Euler a/SGM Uniform. ADetailerとCD Tunerを利用、VAEは導入済み。 基本的には必要ありません。
LINK. TensorArt. yodayo. SeaArt. pixai, x.
(Currently experiencing problems with SeaArt account login and slow uploads SeaArt manager responding. Also, TensorArt is experiencing server errors when uploading models, which also requires time. We are sorry for the inconvenience, but unauthorized reproduction and resale is prohibited. Even if you do, it will be deleted, so it's pointless.)
(現在、SeaArtアカウントのログインに問題があり、アップロードに時間がかかっています。 また、TensorArtではモデルのアップロード時にサーバーエラーが発生しており、こちらも時間がかかっています。 ご迷惑をおかけして申し訳ありませんが、無断複製・転売は禁止されています。見つけ次第マネージャー経由で削除いたします)
How to use this model well
Euler a / SGM Uniform. CFG scale4-6. step20-40. / 720x1280◎. /768x1344, 896×1536/ 1080×1920 /clip skip2.
prompts
masterpiece, best quality, amazing quality, very aesthetic, absurdres, highres, newest, 2023, 2024,Additional Prompts
HDR, 8K, high detailed, subtle, high contrast, detailed background, color depth, chiaroscuro, impasto, smooth shading, glossy surfaces, shiny skin,Also recommended
dak depth, light, low kye, intricate details, masterful shading, detailed background, chromatic aberration, cinematic composition, blurry foreground, blurry background, moody atmosphere, muted colors, etc.Negative prompts
lowres, worst quality, bad quality, bad ???????, jpeg artifacts, watermark, bad finger, missing finger,Models used for merging
coco-style-Illustrious-XLｖ1.0(master-v1.0)
coco-Illustrious-XL(v5.5,v5.6,v6.0,v6.1)
prototype model was created in the process of exploring a new style. It is still rough around the edges, but I promise it will get better.
I am good at coloring, Lighting, and artistic expression. I am also good at portraying people and SFW/NSFW.
I also aim to generate models that do not rely on LoRA or artist tags as much as they should.
And if you like it, please download it, write a review and press the Like icon. All of this will be my motivation.
To prevent unauthorized reproduction and for the benefit of those who wish to build LoRAs, we will be publishing models on other sites in the future. Reproduction of my models on websites by anyone other than myself is prohibited, even if it is not for profit. The sale of any products is prohibited. (01/21/2025)
https://huggingface.co/OnomaAIResearch/Illustrious-xl-early-release-v0
Description
It's been a while! Please check out the new version.
This version is a improved coloring and more faithful to the prompts.
I needed some time to motivate myself. This version is more like a test, so please let us know what you think. We will index the next model.
1. How to use
1) For reforge, turn off Advanced Model Sampling for reForge.
2) CLIP is 2
3) samplers :eular A / SGM Uniform
4) CFG scale4-6. step20-30
5) Observe the recommended settings.
6) No VAE required.
7)Have fun!!
Additional information on usage
Some character tags and Lora may be better with a lower CFG scale, so adjust accordingly. We also recommend adding an animation style to the quality tag.
In some cases, better results may be obtained by using the quality tags Realistic, Photo Background, Photo Reference, etc. without adding anime styles. Several illustrations using these tags have been added since publication.
Recommended setting
Euler a / SGM Uniform. CFG scale4-6. step20-40. 768x1344, 720x1280. clip skip2
prompts
(masterpiece, best quality), amazing quality, very aesthetic, absurdres, highres, newest, HDR, 8K, chiaroscuro, impasto, intrinsic light, luster, shiny color, strong coloring, masterful shading,Negative prompts
lowres, worst quality, bad quality, bad ???????, jpeg artifacts, watermark, nostrils, back light, nose,お久しぶりです！新バージョンをご覧ください。
このバージョンは発色が改良され、よりプロンプトに忠実になっています。
転載問題でモチベーションを上げるのに時間が必要でした。このバージョンはテストのようなものですので、ご意見をお聞かせください。
使い方についての補足
キャラクタータグやロラによっては、CFGのスケールを小さくした方が良い場合がありますので、適宜調整してください。また、クオリティタグにanime stylesを追加することをお勧めします。
また、アニメーションスタイルを追加せずに、クオリティタグにhigh contrast, real, photo background, photo-referencedなどを使用した方が良い結果になる場合もあります。これらのタグを使用したイラストを公開時より複数枚追加しています。