<h3 id="information">Information</h3>This is a Clothing LoRA of an armor set from Skyrim, sourced from Nexus Mods and screenshotted ingame. This is the Anima version. The Illustrious version is <a rel="ugc" href="https://civarchive.com/models/3046187/secret-child-of-talos-or-skyrim-or-illustrious-xl?modelVersionId=3463138">here</a> (once all 20 images aren't flagged for review).<hr /><h3 id="prompt-usage">Prompt Usage</h3>This is an armor set called Secret Child of Talos. It's a two-tone dress with shoulder armor, gauntlets and greaves. Sample Prompt (V1)<pre><code>t4l0sch1ld, 1girl, solo, brown hair, very long hair, single braid, amber eyes, tiara, gorget, cleavage, two-tone dress, short dress, pauldrons, gauntlets, belt, thigh boots, armored boots</code></pre>Sample Prompt (V2)<pre><code>s3cr3tarm0r, 1girl, solo, brown hair, very long hair, single braid, amber eyes, tiara, gorget, cleavage, two-tone dress, short dress, pauldrons, gauntlets, belt, thigh boots, armored boots</code></pre>Prompt Advice <ul><li>Background troubles? Put indoors, stained glass, rug into your negatives (less needed on V2)</li><li>Anime styles? Put 3d, realistic into your negatives (see gallery images).</li></ul><hr /><h3 id="disclaimer">Disclaimer</h3>If you generate imagery with my content, and you'd like to share it, please consider using the <a target="_blank" rel="ugc" href="https://civarchive.com/articles/9229/how-to-give-a-model-creator-50-buzz-at-no-extra-cost-possibly-negative-to-you">"Add Post" button at the bottom of the gallery</a> instead of Post Images button at the top of the site, it gives the creator buzz and it costs nothing on your end. The article that was linked is a very important one I recommend reading. I mainly create for fun, but I would appreciate it regardless.<hr />

Secret Child Of Talos | Skyrim | Anima

V1 is trained at 50 images across 10 checkpoints, 3 repeats, 3 batch size for 1000 steps (30 exposures/image). V1's dataset is all one indoors background. V2 is trained at 100 images across 5 epochs, 5 repeats, 2 batch size for 1250 steps (25 exposures/image). V2's dataset is 50% indoors background, 50% white/grey/black backgrounds, just like the others as of recent, which is how I will do video game armors to fight off the background. V2 might be slightly better but not by a lot. The 25 exposures/image might effectively be 50 since the other 50 images are the same but with a different simple background. Mileage may vary.

Files with SHA256: 5995f012b4963eeb412d8b267e407a19a145a4ad91248ccaccf553cbafde9d51