A high-quality LoRA trained on ~48 carefully curated images, designed to consistently reproduce the same female character across a wide variety of scenarios and styles.Trained on: cyberrealisticPony_v170The dataset includes:<ul><li>multiple angles (frontal, profile, 3/4, low angle, high angle)</li><li>full range of framing (close-up → full body)</li><li>diverse facial expressions and emotions</li><li>varied lighting conditions (soft light, rim light, low-key, neon)</li><li>dynamic poses and motion</li></ul>This results in a LoRA that:<ul><li>maintains strong facial consistency</li><li>generalizes well to new prompts and environments</li><li>avoids overfitting to a single style or background</li></ul>Performs especially well in:<ul><li>realistic photography</li><li>cinematic scenes</li><li>fashion / lifestyle shots</li><li>stylized environments (e.g. cyberpunk, neon lighting)</li></ul><hr />Trigger word: <code>Beka</code> Recommended weight: <code>0.7 – 1.0</code>

Beka

Files with SHA256: 25412b3f0b76e0d2513e5272bbab27ddd37c20391927ea759b7536e4f0dd02e6