<h2 id="jun-park-last-reader-standing-or-character-lora-(sdxl-1.0)">Jun Park — Last Reader Standing | Character LoRA (SDXL 1.0)</h2>A character-focused LoRA that recreates Jun Park, the main hero of the Last Reader Standing universe.<h3 id="what-it-captures">What it captures</h3><ul><li>Face &amp; build – Korean male, mid-20s, messy jet-black hair, deep-brown eyes.</li><li>Signature outfit (<code>JunPark Outfit</code>) – long navy trench-coat, black waist-coat, open white shirt, black compass-rose pendant.</li><li>Casual looks (<code>JunPark Casual Outfit</code>) – linen shirt, yukata, hooded cloak, etc.</li><li>Key props – glowing antique grimoire, golden arcane tattoos/energy, spectral guardian spirits.</li></ul><h3 id="trigger-words">Trigger words</h3><pre><code>JunPark (LRS)
JunParkLRS</code></pre>Add either (or both) at the start of your prompt. Optional outfit tags: <code>JunPark Outfit</code>, <code>JunPark Casual Outfit</code>.<h3 id="recommended-settings">Recommended settings</h3>Setting Value Base checkpoint SDXL 1.0 LoRA weight 0.7 – 0.9 CFG 5 – 7 Clip-skip 1 img2img denoise 0.55 – 0.65<h3 id="prompt-examples">Prompt examples</h3><pre><code>JunPark (LRS), JunPark Outfit, masterpiece, best quality, highly detailed,
korean man channeling golden arcane energy, glowing grimoire, dramatic library lighting</code></pre><pre><code>JunPark (LRS), JunPark Casual Outfit, masterpiece, ultra-detailed,
sunset beach, linen shirt, calm expression, holding ancient book emitting light</code></pre><h3 id="training-details">Training details</h3><ul><li>Dataset : 26 images @ 1024 × 1024 with minimal captions.</li><li>Steps / Epochs : 520 steps (10 epochs).</li><li>Network dim / alpha : 32 / 16 (~55 MB).</li><li>LR : 5 e-4 (UNet) / 5 e-5 (Text Encoder) — cosine scheduler, 3 cycles.</li><li>Other: min-SNR γ = 5, noise offset = 0.1, smart buckets ON.</li></ul><h3 id="license-usage">License / usage</h3>Released under Own Rights: free to use for personal and non-commercial artworks. Commercial use or hateful / defaming content is strictly prohibited.Have fun bringing Jun Park to life! 🚀

Jun Park - Last Reader Standing - Illustrious

Version 1.0 — Initial release (SDXL)What’s new / changed<ul><li>Fresh training from scratch on 26 high-resolution images (1024²) covering Jun’s signature outfit (≈ 80 %) and casual looks (≈ 20 %).</li><li>Base checkpoint: SDXL 1.0.</li><li>Kohya pipeline:<ul><li>network dim 32 / alpha 16 (≈ 55 MB)</li><li>520 steps over 10 epochs</li><li>UNet LR 5 × 10⁻⁴ · Text Encoder LR 5 × 10⁻⁵</li><li>cosine LR scheduler (3 restarts) · min-SNR γ = 5 · noise offset = 0.1</li><li>smart buckets ON, clip-skip 1, batch 4</li></ul></li><li>Trigger words: <code>JunPark (LRS)</code> or <code>JunParkLRS</code><ul><li>Add <code>JunPark Outfit</code> or <code>JunPark Casual Outfit</code> for clothing control.</li></ul></li><li>Recommended weight: 0.7 – 0.9 (txt2img) · denoise 0.55 – 0.65 (img2img).</li><li>Compatible with any SDXL-based checkpoint.</li></ul>Known quirks<ul><li>Very close-up portraits may soften pendant detail; add “sharp pendant” or lower CFG.</li><li>Extreme dynamic poses can introduce minor hand artifacts at strength &gt; 0.9.</li></ul>Roadmap<ul><li>v1.1 (planned) — dim 64 / alpha 32 build with ~50 image dataset for sharper close-ups and broader lighting variety.</li></ul>Enjoy bringing Jun Park to life—feedback is welcome! 🚀

Files with SHA256: 4fb9fd192565df08d79d824d865fe6cbb504b358ff8f8de1e137678e9403cbee