<h3 id="desi-masala-can-be-thought-of-as-the-spiritual-successor-of-this-checkpoint.-find-it-here:-https:civitai.commodels488030desi-masala-xl-nsfw-indiansouth-asian-characters-bxwrmm5rh">Desi Masala can be thought of as the spiritual successor of this Checkpoint. Find it here: <a target="_blank" rel="ugc" href="https://civarchive.com/models/488030/desi-masala-xl-nsfw-indiansouth-asian-characters">https://civarchive.com/models/488030/desi-masala-xl-nsfw-indiansouth-asian-characters</a></h3><h3 id="-bj77f2z49"></h3><hr />This is my first attempt at putting up a model. This was created by first merging many of my LoRAs into <a target="_blank" rel="ugc" href="https://civarchive.com/models/133005?modelVersionId=152076">Juggernaut XL v2</a> , <a target="_blank" rel="ugc" href="https://civarchive.com/models/119229?modelVersionId=149243">ZavyChromaXL v1</a> , <a target="_blank" rel="ugc" href="https://civarchive.com/models/133808/pyros-nsfw-sdxl">Pyro's NSFW SDXL</a>, in various proportions. Following that, the resulting three Checkpoints were merged using A1111.The output does produce nice Indian looks and skin tones, and faces. It has been trained mostly with female images, so the male pics may not come out so great.Suggest using ADetailer followed by upscaling (not hires fix). This is more to fix the flaws of SDXL (cross eyes etc.) rather than anything endemic to the model.This is an alpha version, so please be kind with ratings/ reviews!Do cross post your image generations. would love to see how you use the model.Some of my standard prompts: desi, desilatte, desimocha, desiespresso work well, but aren't required in most scenarios.To-dos:<ol><li>Generating more variations in faces</li><li>Reducing the need for the standard prompts</li><li>Adding more male faces</li><li>Adding Indian backgrounds</li><li>Adding more scaling in skin tone</li></ol>Comments/ reviews welcome!

Files with SHA256: 04b0f14105346930eaf6b9be6e98678aeabea141ef98bf98834c5af19b9b6f88