<h2 id="information">Information</h2><p>This is an experimental LoRA. It gives mixed results.<br />I have found it to 'smooth/refine' Z-Image generations.<br />Please share any results or issues!!<br /></p><h2 id="use">Use</h2><p>No trigger word needed.<br />Use recommended settings.<br />CFG 2-3, steps 20-30<br /><br />Example generations done with: res_multistep/simple/3.0 cfg/18 steps</p><h2 id="training"><br />Training</h2><p>Base Model: <a target="_blank" rel="ugc" href="https://huggingface.co/ostris/Z-Image-De-Turbo/tree/main">Z-Image-De-Turbo </a><br />Dataset: <a target="_blank" rel="ugc" href="https://huggingface.co/datasets/Yejy53/Echo-4o-Image">Echo-4o-Image</a></p>

Files with SHA256: 0025add43e14b432afe97fb45acccdac1953de3b71c390c4b037b00b88d7823c