You can also download a full-merged model here (just in case you don't have Anything V4.5 or want to train a LoRA based on this style): <a target="_blank" rel="ugc" href="https://huggingface.co/OedoSoldier/animix">https://huggingface.co/OedoSoldier/animix</a>你可以在这里下载大模型（以防你没有Anything V4.5或者想要以此画风训练LoRA）：<a target="_blank" rel="ugc" href="https://huggingface.co/OedoSoldier/animix">https://huggingface.co/OedoSoldier/animix</a><h3>Check the new variant: <a rel="ugc" href="https://civarchive.com/models/26622">Ambientmix</a>!</h3>Recommend settings:Model: Anything V4.5VAE: Orangemix (the same with NAI)LoRA Strength: 1Sampler: DPM++ 2M KarrasSampling steps: 20CFG: 7Negative embedding: <a target="_blank" rel="ugc" href="https://civarchive.com/models/7808/easynegative">EasyNegative</a>、<a target="_blank" rel="ugc" href="https://civarchive.com/models/16993/badhandv4-animeillustdiffusion">badhandv4</a>Highres fix is also recommended.Note: all the LoRA names used in sample images are my local name, you need to change them to your saved LoRA filename!推荐设置：基模：Anything V4.5VAE：Orangemix（同NAI）LoRA强度：1采样器：DPM++ 2M Karras采样步数：20CFG：7负面embedding：<a target="_blank" rel="ugc" href="https://civarchive.com/models/7808/easynegative">EasyNegative</a>、<a target="_blank" rel="ugc" href="https://civarchive.com/models/16993/badhandv4-animeillustdiffusion">badhandv4</a>推荐使用高清修复。注意：样图中的LoRA名称是我本地文件名，要改成你保存的文件名！This model is suitable for reproducing images similar to anime screenshots, and it is recommended to use <a target="_blank" rel="ugc" href="https://github.com/hako-mikan/sd-webui-lora-block-weight">LoRA block weight</a> expansion when using LoRA and character LoRA together. Setting the block weights of the character LoRA to presetting <code>MIDD</code> will reduce the distortion caused by the influence of the character LoRA on the style.本模型适合复现类似于动漫截图的图像，建议在此LoRA与角色LoRA一起使用时使用<a target="_blank" rel="ugc" href="https://github.com/hako-mikan/sd-webui-lora-block-weight">LoRA块权重扩展</a>。将角色LoRA的输出权重设置为<code>MIDD</code>预设将减少角色LoRA影响对风格产生的影响。 This is a LoRA mix of:<ul><li>2020s Anime Magazine Illustration Style</li></ul><ul><li>Anime-like 2D (extracted LoRA)</li><li>Anime Lineart Style</li><li>Anime Screencap Style</li><li>Avas Anime Hamster</li><li>Epi Noise Offset</li><li>Hipoly 3D Model Lora</li><li>Makoto Shinkai Substyles</li></ul>Merged with special tweaks to make full use of the strengths of each LoRA.本LoRA由以下LoRA混合而成：<ul><li>2020s Anime Magazine Illustration Style</li></ul><ul><li>Anime-like 2D (extracted LoRA)</li><li>Anime Lineart Style</li><li>Anime Screencap Style</li><li>Avas Anime Hamster</li><li>Epi Noise Offset</li><li>Hipoly 3D Model Lora</li><li>Makoto Shinkai Substyles</li></ul>经过特别调教，得以发挥各LoRA的长处。

Animix - Anime Screenshot-like Style Mix LoRA (アニメスクショ風/动画截图风)

Just in case you don't have Anything V4.5 or want to train a LoRA based on this style

Files with SHA256: ad0e54efe2348cffc147dc3e80361a3e2fa0eb094585e0495d7a499399a7d8b7