Trained on 195 HD images with heavily curated tags of Arthur Rackham's golden age illustrations.<ul><li>ALL TAGS -&gt; <a target="_blank" rel="ugc" href="https://drive.google.com/file/d/1GhiFrsNXsHN_GZOVaLXFzUkbjss1eOqM/view?usp=sharing">GDrive</a> &lt;-</li><li>Booru-style prompts - (e.g., 1girl, 1boy, old man, size difference, age difference, giant, wrinkled skin, long nose, nymph, gnome, faux scenery, etc.)</li><li>When prompting for two+ subjects try adding a slash between the subjects: 1girl/1boy</li><li>Depending on the style you want (ink, watercolor, etc.), it helps to include the desired medium in the prompt and place any unwanted mediums in the negative prompt. For example - Positive: watercolor_(medium) Negative: ink_(medium) That said, it's not a strict rule! Some mediums can work well together, so feel free to experiment~</li></ul><hr />✨ If you enjoy my LoRAs and creations, you can support me with a little coffee on <a target="_blank" rel="ugc" href="https://ko-fi.com/rixyn">Ko-fi</a> ☕💖It helps me keep training models, creating art and making more dreamy weird little projects 🌙✨ Thank you so much for the support and kindness ~<hr />SD1.5 VERSION Trained on 235 HD/upscaled images of Arthur Rackham's work.<ul><li>STRENGTH/WEIGHT: 0.6 - 1.2</li><li>Trained on base Stable Diffusion 1.5 (with anime/booru prompts; Yet flexible)</li><li>You might have to shout at it for specific tags, for example: (old, man, wrinkled skin:1.6) &lt;- depending on the model and strength of the lora</li><li>Generates humans sometimes when asked not to.. so add a higher weight to negative prompts for 1girl, 1boy, human, etc.</li><li>Wild west tagging (auto-tagged)</li><li>Activation tag is a mash-up of "fairytale" and "illustration" with a number instead of an "i" - fa1l</li></ul>

Files with SHA256: ba5b70cc3196450516ccd39a72212d76e5f12f358f8ed91c2f6edfc840a93ee1