Initialized from OpenJourney v2, further fine-tuned for 4000 steps on images scraped from the front page of Lexica art (January 2023). Good at producing "cinematic 3D" looking images. To trigger it best, use prompts that look like things people on the front page of <a target="_blank" rel="ugc" href="http://Lexica.art">Lexica.art</a> would do, like "sharp focus", "artgerm", "unreal engine", "greg rutkowski", "wlop", "trending on artstation", etc. Simple example prompt: "Techno magic robot, side portrait, realistic, half body shot, sharp focus, elegant" Long example prompt: "wild gremlin woman with green hair, hyper detailed, digital art, cyberpunk style, cinematic lighting, studio quality, smooth render, unreal engine 5 rendered, octane rendered, art style by klimt and nixeu and ian sprigger and wlop and krenz cushart" (Disclaimer: I have no idea what half the text of this prompt is meant to be, it's just yoinked from the stuff lexica users type)

Files with SHA256: f2bb36eb2e5412852c78453948acdbf985c7127f73731e75cb123d583a62c56a