Trained on vanilla SD1.5 but works better with art-focused checkpoints like Elldreth's and others from this website. Training dataset contained game portraits, key art, and some cherrypicked fanart.How to consistently get very stylized results:<ul><li>Use Elldreth's lucid mix</li><li>Use "[{subject}:abstract background:0.5], expressionism, digital art, colorful background" prompt, replace {subject} with anything</li><li>Parameters: 20 steps, CFG=2, DPM++ 2M Karras sampler</li><li>Some usual negative prompts</li></ul>See my comment below for examples of this method

Files with SHA256: 4fa93321d0410cd2c602df983901624faa33a263b3bc2f057826fe2059d75230