Used several models available on Civitai to make this merge all credit goes to their authors. Feel free to improve this. Merge of Next models <a target="_blank" rel="ugc" href="https://civarchive.com/models/1246404/wai-fxxkill-xl">https://civarchive.com/models/1246404/wai-fxxkill-xl</a> <a target="_blank" rel="ugc" href="https://civarchive.com/models/1318945/one-obsession">https://civarchive.com/models/1318945/one-obsession</a> v18 <a target="_blank" rel="ugc" href="https://civarchive.com/models/1034922/dasiwa-illustrious-xl">https://civarchive.com/models/1034922/dasiwa-illustrious-xl </a>(sweet poison ver) <a target="_blank" rel="ugc" href="https://civarchive.com/models/1325426/amanatsu-illustrious">https://civarchive.com/models/1325426/amanatsu-illustrious </a>(v2) <a target="_blank" rel="ugc" href="https://civarchive.com/models/1203050/fabricated-xl">https://civarchive.com/models/1203050/fabricated-xl </a>v6 Was experimenting to be good with DMD2 Lora 1.0 and Euler Karass and optionally TokenMerging 0.30. Due to excessive merging fingers and eyes are not sharp. Recipe 1st Merge <img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/596a0cc4-47e5-41ba-b4af-45eb26986e2a/width=525/596a0cc4-47e5-41ba-b4af-45eb26986e2a.jpeg" />2nd Merge <img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/7401c1d1-c6b8-4359-a871-8c37d49eb485/width=525/7401c1d1-c6b8-4359-a871-8c37d49eb485.jpeg" />3rd Merge <img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/07e612a2-7d95-4504-935c-8f710c00c608/width=525/07e612a2-7d95-4504-935c-8f710c00c608.jpeg" />4th Merge <img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/5cb6ae80-8daf-4387-8ed6-80538d941d4f/width=525/5cb6ae80-8daf-4387-8ed6-80538d941d4f.jpeg" />V1.0_1 Unet part wasn't touched. There my experiment stops. Better to modify workflow than trying to adapt style of a model to existing one. Tried to merge different clips to fix distant faces problem, and stopped at the one mentioned in model info at ratio 0.7 it somewhat fixed it in my workflow, but in simple workflows it is still the same (or worse?), While generating noticed that V1.0 and V1.0_1 is very bad with "<span style="background-color:rgb(17, 17, 17);color:rgb(238, 238, 238);font-family:SFMono-Regular, Menlo, Monaco, Consolas, &quot;Liberation Mono&quot;, &quot;Lucida Console&quot;, &quot;Courier New&quot;, Courier, monospace, monospace;font-size:16px">full body" keep that in mind better use <span style="background-color:rgb(17, 17, 17);color:rgb(238, 238, 238);font-family:SFMono-Regular, Menlo, Monaco, Consolas, &quot;Liberation Mono&quot;, &quot;Lucida Console&quot;, &quot;Courier New&quot;, Courier, monospace, monospace;font-size:16px">"full-length portrait". <img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/ed6daa0f-1c48-4cdf-8b1e-7230ff8784f3/width=525/ed6daa0f-1c48-4cdf-8b1e-7230ff8784f3.jpeg" />

SweetBread

Merged CLIP from v1 into Licentious Anime CLIP <a target="_blank" rel="ugc" href="https://civitai.com/models/2145238/licentious-anime-illust?modelVersionId=2426459">https://civitai.com/models/2145238/licentious-anime-illust?modelVersionId=2426459</a> at ratio 0.70, to get better results with DMD2.

Files with SHA256: 7393808efdb07f1939051bb65ec61c7d2eb78cb855a694dc32b45b87f521ca96