HosekiMx

vaeKlF8Anime2_klF8Anime2VAE.safetensors

hosekimix_v10.safetensors

v1.0

If you want to support me, consider liking the model and tipping buzz!A mix used for my LoRA previews, I wanted a good middle ground between color vibrance and details / shading of some models I like.<h2 id="heading-101">Parameters I use :</h2><ul><li>Positive prompt : <code>masterpiece, best quality</code></li><li>Negative prompt : <code>(worst quality, low quality:1.4), signature, watermark, username, blurry</code></li><li>Sampler : DPM++2M Karras - 25 steps</li><li>CFG Scale : 5</li><li>Hiresfix : 4x-AnimeSharp-lite - 0.4 denoising</li><li>Clip Skip : 2</li><li>VAE : <a target="_blank" rel="ugc" href="https://civarchive.com/models/23906/kl-f8-anime2-vae">kl-f8-anime2</a></li></ul>The VAE is not baked in, so you can download it or use another one if you'd prefer.I also used the <a target="_blank" rel="ugc" href="https://civarchive.com/models/7808/easynegative">EasyNegative</a> embedding for the negative prompt, and <a target="_blank" rel="ugc" href="https://github.com/Bing-su/adetailer">adetailer</a> for more detailed faces.The models in the merge are <a target="_blank" rel="ugc" href="https://civarchive.com/models/6174?modelVersionId=122702">DivineEleganceMix</a> (v6), <a target="_blank" rel="ugc" href="https://civarchive.com/models/9139?modelVersionId=11086">YesMix</a> (v1.5), <a target="_blank" rel="ugc" href="https://civarchive.com/models/53360?modelVersionId=106169">GalenaREDUX</a> (v2), <a target="_blank" rel="ugc" href="https://civarchive.com/models/116279/seizamix">seizaMix</a>, <a target="_blank" rel="ugc" href="https://civarchive.com/models/34553?modelVersionId=150207">AingDiffusion</a> (v9), and <a target="_blank" rel="ugc" href="https://civarchive.com/models/85909?modelVersionId=91337">Sudachi</a>.This is my first merge, let me know if there's any problems!<h3 id="heading-2">Hope you enjoy!</h3>

HosekiMix

creator: HosekiA mix used for my LoRA previews, I wanted a good middle ground between color vibrancy and details / shading of some models I like.Parameters I use&nbsp;:<ul><li>Positive prompt :&nbsp;masterpiece, best quality</li><li>Negative prompt :&nbsp;(worst quality, low quality:1.4), signature, watermark, username, blurry</li><li>Sampler : DPM++2M Karras - 25 steps</li><li>CFG Scale : 5</li><li>Hiresfix : 4x-AnimeSharp-lite - 0.4 denoising</li><li>Clip Skip : 2</li><li>VAE :&nbsp;<a href="https://civitai.com/models/23906/kl-f8-anime2-vae" rel="ugc noopener noreferrer nofollow" target="_blank">kl-f8-anime2</a></li></ul>The VAE is&nbsp;not baked in, so you can download it or use another one if you'd prefer.I also used the&nbsp;<a href="https://civitai.com/models/7808/easynegative" rel="ugc noopener noreferrer nofollow" target="_blank">EasyNegative</a>&nbsp;embedding for the negative prompt, and&nbsp;<a href="https://github.com/Bing-su/adetailer" rel="ugc noopener noreferrer nofollow" target="_blank">adetailer</a>&nbsp;for more detailed faces.The models in the merge are&nbsp;<a href="https://civitai.com/models/6174?modelVersionId=122702" rel="ugc noopener noreferrer nofollow" target="_blank">DivineEleganceMix</a>&nbsp;(v6),&nbsp;<a href="https://civitai.com/models/9139?modelVersionId=11086" rel="ugc noopener noreferrer nofollow" target="_blank">YesMix</a>&nbsp;(v1.5),&nbsp;<a href="https://civitai.com/models/53360?modelVersionId=106169" rel="ugc noopener noreferrer nofollow" target="_blank">GalenaREDUX</a>&nbsp;(v2),&nbsp;<a href="https://civitai.com/models/116279/seizamix" rel="ugc noopener noreferrer nofollow" target="_blank">seizaMix</a>,&nbsp;<a href="https://civitai.com/models/34553?modelVersionId=150207" rel="ugc noopener noreferrer nofollow" target="_blank">AingDiffusion</a>&nbsp;(v9), and&nbsp;<a href="https://civitai.com/models/85909?modelVersionId=91337" rel="ugc noopener noreferrer nofollow" target="_blank">Sudachi</a>.This is my first merge, so I'm probably going to update this at some point, let me know if there's any problems!

Sampler: DPM++ 2M Karras

Negative: EasyNegative, (worst quality, low quality:1.4), signature, watermark, username, blurry

Creator : <a target="_blank" rel="noopener noreferrer nofollow" href="https://civitai.com/user/Hoseki">Hoseki</a><a target="_blank" rel="noopener noreferrer nofollow" href="https://civitai.com/user/Hoseki">Par</a>ameters&nbsp;:<ul><li>Positive prompt :&nbsp;masterpiece, best quality</li><li>Negative prompt :&nbsp;(worst quality, low quality:1.4), signature, watermark, username, blurry</li><li>Sampler : DPM++2M Karras - 25 steps</li><li>CFG Scale : 5</li><li>Hiresfix : 4x-AnimeSharp-lite - 0.4 denoising</li><li>Clip Skip : 2</li><li>VAE :&nbsp;kl-f8-anime2</li></ul>The VAE is&nbsp;not baked in, so you can download it or use another one if you'd preferAlso used the&nbsp;EasyNegative&nbsp;embedding for the negative prompt, and&nbsp;adetailer&nbsp;for more detailed facesThe models in the merge are&nbsp;:DivineEleganceMix&nbsp;(v6),&nbsp;YesMix&nbsp;(v1.5),&nbsp;GalenaREDUX&nbsp;(v2),&nbsp;seizaMix,&nbsp;AingDiffusion&nbsp;(v9), and&nbsp;Sudachi

Files with SHA256: 21021753bd1ba0b094e217f7f7d0608d4720339c05562a60a69998d39cb9a11c