My attempt in creating an SDXL Checkpoint merge that require almost no loras to achieve any desired output.The main focus is to achieve a model that can give excellent results at CFG 1. If you like my work you can buy me a <a target="_blank" rel="ugc" href="https://ko-fi.com/knos94">coffee</a>. Thanks💖Main Features:<ul><li>Works with a wide variety of prompts.</li><li>Extremely good prompt adherence especially with text.</li><li>Even at CFG 1 the text adherence is good.</li></ul>Recommended Generation Parameters:<ul><li>Sampler: LCM</li><li>Scheduler: Karras/Exponential/beta57</li><li>Steps: 14 -18</li><li>CFG: 1-2 (Don't recommend going above 1.8)</li></ul>Models used for the merge mentioned in suggested resources.Hope you guys like the model. Eager for the feedback..

Pikon-Realism

Merged it with few more of my favourite models (mentioned in suggested resources). After a few merges i ended up with two models with varying styles; v2 and v2(alt).

Files with SHA256: c720b21a56cf39d918ffc71aa013f764b5139a381afa0a691c131aceca302eed