<h1 id="ranka-ookami-lora-murenase!-seton-gakuen">Ranka Ookami LoRA -= Murenase! Seton Gakuen</h1><p>Trained on Anima preview 3 (I recommend using AnimaYume V0.4) · Danbooru tags + very minimal Natural Language (noticed some issues with the dataset after training but it should be mostly fine)</p><p>A LoRA for Ranka Ookami (大神ランカ) from Murenase! Seton Gakuen<br /></p><p><strong>Recommended Base Prompt:</strong></p><pre><code>masterpiece, best quality, score_7,</code></pre><p>additionally score_8, score_9,</p><pre><code>1girl, solo, ranka ookami, wolf girl, </code></pre><pre><code>animal ears, wolf ears, animal ear fluff, large fluffy pink tail, </code></pre><pre><code>bright pink hair, long hair, flat chest.</code></pre><h2 id="training-details">Training details</h2><p><strong>Base model: Anima 3 Preview</strong></p><p>Dataset: ~81 images - anime screenshots</p><p>Caption style: Mixed (danbooru tags + Min NL descriptions)<br /></p><h2 id="tips-and-known-issueslimitations">Tips and Known issues/Limitations</h2><p>Use wolf ears, girl, etc... tags explicitly - the dataset has some cat ears / fox ears tag noise (woopsie - old dataset merged with new one 😝)</p><p>NL prompting may be less consistent than tag prompting due to the dataset caption style</p><p></p>

Files with SHA256: aedade886a0520521c6535a9872c48f2ec398cad19d559afca12957cb7e4cc64