an illustrative style transfer lora trained on varying mediums, with loose brushstroke imagery being the primary focus. adding tags <code>sketch</code>, <code>monochrome</code>, <code>lineart</code>, and <code>greyscale</code> to your prompt switches it to a looser greyscale sketch style for use with <a target="_blank" rel="ugc" href="https://huggingface.co/kohya-ss/Anima-LLLite">lllite</a>. while the samples are focused on furry subjects: human characters, anime girls, scenery, and food should all pick up the style<ul><li>the trigger word for activating this lora is <code>@dededo</code></li><li>for captioning, booru-style tags or natural language both work. please note that tag dropout was not used, so you may need to be specific in your prompting when tagging</li><li>my workflow is embedded into the sample images, feel free to check it out if you need help with generation</li><li>trained on anima base with <a target="_blank" rel="ugc" href="https://github.com/tdrussell/diffusion-pipe">diffusion pipe</a> and cuda 13.2/pytorch nightly</li><li>settings i use on average, entirely optional—<ul><li>strength: 1.0 ; <a target="_blank" rel="ugc" href="https://docs.comfy.org/built-in-nodes/CreateHookKeyframe">start percent</a>: 0.3</li><li>settings: er_sde, simple/<a target="_blank" rel="ugc" href="https://github.com/ClownsharkBatwing/RES4LYF">beta57</a> ; cfg 5.0, 30 steps</li></ul></li></ul><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/4ade4c0e-f5a8-4d6e-8ce2-3735f498bfaf/original=true/4ade4c0e-f5a8-4d6e-8ce2-3735f498bfaf.jpeg" />

meow-v1.1-epoch12.safetensors

meow-v1.1-epoch10.safetensors

meow-v1.1-definitive.safetensors

Files with SHA256: ba4b119d1337021abdf91734ae3400e7e6fc6c5b201a936f1ae006fb951ea306