High-fidelity character LoRA for Senko-san from The Helpful Fox Senko-san, for use with Boogu-Image-0.1 Base.Built from 1,000 native 1024 × 1024 NovelAI 5.0 Curated generations, this LoRA targets sharp anime images with defined linework, solid anatomy, and minimal compression blur. It is designed for high-quality <code>txt2img</code> character generation and <code>img2img</code> with strong identity consistency and flexible natural-language prompting.This ULTRA version combines the 598-image base dataset used to make my NovelAI 4.5 Curated LoRAs with 402 new activity scenes, expanding scene and activity versatility while preserving Senko’s likeness.<h2 id="the-trigger-word">The Trigger Word</h2><pre><code>senko_san</code></pre>Include <code>senko_san</code> in your positive prompt, followed by a description of the outfit, activity, setting, and visual style you want. Use the Base model. Turbo produces poor results with this LoRA and is not recommended.Character Features<ul><li>Recognizable identity: Senko’s face, amber eyes, short layered golden-blonde hair, fox ears, fluffy fox tail, and hair ornament.</li><li>Outfit flexibility: Traditional clothing, kimono, dresses, casual outfits, jackets, hoodies, and other wardrobe choices.</li><li>Expressive scenes: Varied poses, expressions, activities, and compositions.</li><li>Flexible environments: Indoor and outdoor scenes, traditional and modern settings, and different times of day.</li></ul>PromptingUse descriptive natural language. Start with the trigger word, then explain the image in clear sentences.<ul><li>Describe her clothing and expression explicitly.</li><li>Specify the pose or activity you want.</li><li>Include the setting, framing, and lighting.</li><li>State the intended style, such as anime illustration, clean linework, or soft cel shading.</li><li>Keep instructions consistent and prioritize the details that matter most.</li></ul>Model Setup — ComfyUIUse an updated ComfyUI installation with native Boogu support.<ul><li>Base model: <code>Boogu-Image-0.1-Base</code></li><li>Text encoder: Qwen3-VL-8B, such as <code>qwen3vl_8b_fp8_scaled.safetensors</code></li><li>Text encoder loader type: <code>boogu</code></li><li>VAE: FLUX.1 VAE, such as <code>flux1_vae_bf16.safetensors</code></li><li>Trigger word: <code>senko_san</code></li></ul>Suggested Starting Settings<ul><li>LoRA weight: Start at <code>1.0</code></li><li>Sampler: Euler (<code>euler</code>)</li><li>Scheduler: Simple (<code>simple</code>)</li><li>Steps: Start at <code>30</code>; explore <code>25–50</code></li><li>CFG scale: Start at <code>4.0</code>; explore <code>2.0–5.0</code></li><li>Resolution: Start at <code>1024 × 1024</code></li><li>Denoise: <code>1.0</code> for <code>txt2img</code></li><li>Negative prompt: Start empty; add targeted exclusions if needed</li></ul>Adjusting LoRA StrengthStart at <code>1.0</code> and adjust in small increments.<ul><li>Try <code>0.85–1.0</code> to balance character identity with outfit and style flexibility.</li><li>Lower the weight if character details overpower the requested clothing or composition.</li><li>Try <code>1.1–1.2</code> for stronger character influence, checking for exaggerated features or reduced flexibility.</li></ul>

Senko-san_NAI-V5-Curated_HD_ULTRA (仙狐さん)

NovelAI 5.0 Curated Dataset - 1,000 HD Images

Files with SHA256: 131b87afab9476ade3f67f32a7f03de0a1eafa7b2bd1b0c07b0ef10727b5f725