he model is aimed at photorealism and trying for a consistency with formRecommended Negative Prompt:(worst quality, low quality, illustration, 3d, 2d, painting, cartoons, sketch), open mouthor another negative promptᅠRecommended Generation Parameters: Sampling Steps: 15-30+ Sampling Method: DPM++ 2M Karras / or anotherLora: &lt;1.3/1.35 of 2.0 has seemed bestFor Checkpoints, so far I've found most success using:<a target="_blank" rel="ugc" href="https://civarchive.com/models/277058?modelVersionId=646523">epiCRealism XL</a><a target="_blank" rel="ugc" href="https://civarchive.com/models/139562?modelVersionId=344487">RealVisXL V4.0</a>As per requested, no name, Nor will any permissions be given for use in commerce, monetization, or nsfw applications.

Baraan XL

Anna-000010.safetensors

1.1v

So far:<ul><li>0.85-1.25 of 2.0, appears fairly consistent, (1.15 or there about's may be best), up in the higher weight mixing lora had more troubles.-Hair length doesn't always follow prompt, esp. w/ pixie or between bald and bob</li><li>Mixed bag with playing nice with other lora, results are mixed, if with one and neither is set too high mostly ok, things start to fall apart with two or more, </li><li>As for steps I'm thinking 30+, it probably is fine down to 20, over 39/40 things over-processed. </li></ul>

Files with SHA256: cd5ed0a07b51840562ac67194526103a3c9d23602e7e1fb9f9ba8f4608b54e1e