<a target="_blank" rel="ugc" href="https://www.patreon.com/cw/Techannel"><img src="https://cdn.jsdelivr.net/gh/Techannel/assets@v3/banners/patreon-banner.png" alt="Early access on Patreon: all early-access LoRAs free for 7 days, then €5/month / 早期アクセスのLoRAすべて、7日間無料でお試し" /></a><a target="_blank" rel="ugc" href="https://techannel.app/"><img src="https://cdn.jsdelivr.net/gh/Techannel/assets@v3/banners/archive-banner.png" alt="Techannel archive: see what is in early access and vote on the next LoRA, free / 早期アクセス中のLoRAを見る・次のLoRAに無料で投票" /></a>💬 Discord — <a target="_blank" rel="ugc" href="https://discord.gg/TYHy7fpEXx">join here / 参加はこちら</a> · 🔓 Patreon — <a rel="ugc" href="https://www.patreon.com/cw/Techannel">new LoRAs 7 days early / 新作LoRAを7日早く</a> · ☕ Ko-fi — <a target="_blank" rel="ugc" href="https://ko-fi.com/techannel">support the factory / 応援する</a> · 💼 <a rel="ugc" href="https://ko-fi.com/techannel/commissions">Request Commission / コミッション依頼</a><h1 id="ronova-genshin-impact">Ronova ロノヴァ — Genshin Impact 原神</h1>The horns bake, the veil bakes. the wings just show up.<hr />Character LoRA for Ronova, the Ruler of Death in Genshin Impact. The trigger <code>ronovagenshin</code> bakes in what her captions never named: her face, the very long platinum-white hair, its intakes and sidelocks, the gold eyes, the diamond flower pupils, the red horn ornaments, the veil, and the white fur mantle.Her outfit stayed tagged, so it stays steerable: the <code>black dress</code>, the <code>black bodysuit</code> and <code>bodystocking</code>, <code>black gloves</code>, <code>elbow gloves</code>, <code>bare shoulders</code>, <code>pantyhose</code>, and <code>earrings</code>. Prompt them for canon, or leave them out. the face holds either way.And the flourish: her wings arrive nearly unprompted. the red span with the eye-marks down it, turning up on a bare <code>ronovagenshin, 1girl, solo</code>. Reach for <code>red wings</code> and <code>extra eyes</code> if a seed is shy.<h2 id="version">🧩 Version</h2><ul><li>🔵 v2.0 Anima, for Anima / Qwen-Image DiT checkpoints. ✅ Works in the Civitai on-site generator. Weight 1.0 (usable range 0.85 to 1.0). Same trigger, stronger likeness, the star pupils finally read, and the wings come on their own.</li><li>🔵 v1.0 Anima for Anima / Qwen-Image DiT checkpoints. ✅ Works in the Civitai on-site generator. Weight 1.0 (usable range 0.85 to 1.0).</li><li>🟢 v1.0 Illustrious for SDXL / Illustrious / Pony / NoobAI checkpoints. Weight 1.0 (usable range 0.85 to 1.0).</li><li>🌸 v1.0 Krea 2 for Krea 2 checkpoints (Raw &amp; Turbo). Same trigger <code>ronovagenshin</code>. Weight 1.0. Prompts in natural language. see the Krea 2 section.</li><li>🟠 Remapped Anima 2.9B — this version is the same LoRA remapped onto the 40-layer Anima-2.9B layout (weight 0.5–0.7, no patch node needed). It is a remap, not a retrain, so there may be issues with the 2.9B version — if there is enough interest I may retrain it natively for 2.9B.</li></ul><blockquote>v2.0 Anima is the file to grab. The older files and their prompts stay below.</blockquote><h2 id="prompt-v2">📋 Prompt, v2.0 Anima</h2>Canon look, wings and all:<pre><code>ronovagenshin, wings, red wings, veil, black dress, black gloves</code></pre>Negative, keep it:<pre><code>3d, cel shading, render, chibi, text, character name, multiple girls, simple background, white background</code></pre><h2 id="prompt">📋 Prompt, v1.0 files (old prompts)</h2>Full canon kit:<pre><code>ronovagenshin, black dress, black bodysuit, black gloves, bare shoulders, pantyhose, halo</code></pre>Short stack, identity only:<pre><code>ronovagenshin</code></pre>Winged, the eye-marked span, pushed:<pre><code>ronovagenshin, (wings:1.2), red wings, extra eyes</code></pre>Negative, keep this one:<pre><code>chibi, english text, text, logo, watermark, character name, reference sheet, signature</code></pre><h2 id="krea-2">🌸 Krea 2</h2>The Krea version speaks prose, not tags, on the same trigger <code>ronovagenshin</code>. the face, platinum hair, gold eyes, horn ornaments, veil and fur mantle all ride the trigger. Describe the outfit and the wings in plain language:<pre><code>ronovagenshin, full body shot of the subject standing in a black ornate dress with red accents, large red wings marked with eye patterns spread behind her</code></pre>No negative needed. the tag-lane negative above is for Anima / Illustrious only.<h2 id="tips">💡 Tips</h2><ul><li>Her eyes need pixels. The amber star pupil is in the LoRA, but at small face sizes any model smears it. Use a face detailer or a hi-res pass on portraits and it comes back. Close-ups with <code>eye focus</code> show it best.</li><li>On v2.0 the wings are always there. They are her signature and the training pool agreed, so a negative will not remove them. A full outfit change does: describe the clothes and drop <code>wings</code> from the prompt.</li><li>Keep <code>3d, cel shading, render</code> in the negative on v2.0. Part of its training set was cel-shaded, that line keeps the output painted.</li><li>That negative is not optional. A lean prompt slides into a game-card layout: chibi figure, framed panel, her name underneath.</li><li>Never put her hair colour in the negative. The platinum is baked, so <code>white hair</code> as a negative fights the trigger. Coloured ambient light tints it. Light neutral, or take the cast.</li><li>Gold eyes ride the trigger. If fan-art variance pulls them red or blue, negative <code>red eyes, blue eyes</code>.</li><li><code>pointy ears</code> is promptable. It needs asking.</li><li>Her halo is trained: <code>halo</code>, or <code>mechanical halo</code>.</li><li>For alt outfits, negate the canon one. Put <code>black dress, black bodysuit</code> in the negative and she takes a coat, a <code>hood</code>, or <code>armor</code>.</li><li>Pool skews to group art. keep <code>solo</code> in.</li></ul><hr />More of my LoRAs: <a target="_blank" rel="ugc" href="https://civarchive.com/user/Techannel">Techannel on Civitai</a>. Likes &amp; Buzz keep the factory running ❤️

Ronova ロノヴァ — Genshin Impact 原神 [Anima+Illustrious]

🟠 Remapped Anima 3.8B — this version is the same LoRA remapped onto the 52-layer Anima 3.8B / Delta Mix layout (weight 0.65–1.0, plain Load LoRA, no patch node needed). It is a remap, not a retrain, so there may be issues with the 3.8B version.

Files with SHA256: a6d717ab53795380547d202097e0be020319d0f34626bc42f7e4dc5506e240fe