<h1 id="premium-character-loras">Premium character LoRAs</h1>Original synthetic characters, bundles and AI tools are in my shop:<a target="_blank" rel="ugc" href="https://rkrra.com/civitai">https://rkrra.com/civitai</a>Get the character LoRAs on Patreon: <a target="_blank" rel="ugc" href="https://www.patreon.com/cw/RkRRa">https://www.patreon.com/cw/RkRRa</a>Join the community on Discord: <a target="_blank" rel="ugc" href="https://discord.gg/nB79wRNtMT">https://discord.gg/nB79wRNtMT</a><h2>Andorian Male 1</h2>A character LoRA for Krea 2. Reproduces a consistent subject across poses, framing, and lighting.<h3>Usage</h3>Include the trigger word <code>andrnml1</code> in your prompt.Recommended weight: 0.7. At weight 1 this LoRA also pulls other people in the image towards this character: ask for someone else and you may still get this face. At 0.7 the likeness holds and the rest of your prompt is respected. If you need a stronger likeness, go up in small steps.Two character LoRAs at once tend to blend, especially when the two characters look alike. For two characters in one image, generate them one at a time.<h3>Prompt Examples</h3>Here are some prompt examples:<blockquote>"andrnml1, chest-up, straight-on, drinking from glass, neutral expression, leather jacket over tee, plain gray backdrop, soft even lighting."</blockquote><blockquote>"andrnml1, chest-up, eye-level, leaning on counter, smiling, urban night street, neon signs glowing, wet pavement reflecting lights, pedestrians blurred in background, warm ambient glow from street lamps and storefronts."</blockquote><blockquote>"andrnml1, chest-up, eye-level, standing upright, gazing forward, neutral expression, industrial shipping containers backdrop, warm golden hour light, soft shadows, outdoor setting, stacked cargo visible behind."</blockquote><blockquote>"andrnml1, three-quarter from mid-thigh up, straight-on, kneeling with hands clasped, stern expression, black tuxedo with bow tie, dark blue backdrop, studio lighting, formal posture, alien antennae visible, composed demeanor, elegant attire, minimalistic setting."</blockquote><blockquote>"andrnml1, chest-up, eye-level, shouting with mouth open, animated gesture, public transit station, bright overhead lighting, blurred commuters, digital display boards, modern architecture."</blockquote><blockquote>"andrnml1, chest-up, straight-on, standing upright, gazing forward, neutral expression, wooden cabin interior, warm ambient lighting from windows and lamps, fireplace glow, framed artwork on walls."</blockquote><blockquote>"andrnml1, close-up portrait, eye-level, facing forward, slight smile, outdoor golf course, soft warm light, blurred greenery and hills, calm demeanor, natural daylight, tranquil setting, peaceful expression."</blockquote><blockquote>"andrnml1, full length from behind, straight-on, walking away, hands relaxed at sides, dark cable-knit sweater, dark pants, deep red studio backdrop, soft even lighting from upper left, large octagonal softbox visible in corner."</blockquote><blockquote>"andrnml1, chest-up, slightly low angle, head turned right, smiling subtly, studio with draped backdrop and window light, soft warm illumination from side."</blockquote><blockquote>"andrnml1, chest-up, eye-level, standing still, stern expression, dark coat, knit scarf, residential street at night, warm artificial lights from windows, cool ambient glow from street lamps."</blockquote><blockquote>"andrnml1, chest-up, eye-level, arms wide open, smiling, urban night street, warm glowing streetlights, fountains, pedestrians, buildings, soft ambient glow, festive lights, inviting stance."</blockquote><blockquote>"andrnml1, three-quarter from mid-thigh up, slightly low angle, mid-stride with arms bent, smiling, professional kitchen with flames and chefs, warm directional lighting from left, stainless steel surfaces, vibrant vegetables."</blockquote><blockquote>"andrnml1, chest-up, straight-on, seated writing, focused expression, dark plain background, soft directional lighting, hands holding pen over open book, formal posture, quiet concentration, professional demeanor."</blockquote><blockquote>"andrnml1, full length from behind, low angle, reaching upward toward shelf, neutral expression, grand library with wooden cabinets and arched window, soft natural light filtering through high window."</blockquote><blockquote>"andrnml1, chest-up, straight-on, facing forward, neutral expression, greenhouse with plants, natural diffused light, glass roof structure, soft shadows, indoor botanical environment, structured framing."</blockquote><blockquote>"andrnml1, chest-up, eye-level, shouting with mouth open, outdoor bus stop, bright daylight, brick buildings, parked cars, clear blue sky, metal frame structure, visible timetable board, natural sunlight casting soft shadows."</blockquote><blockquote>"andrnml1, chest-up, eye-level, holding stacked books, smiling, urban plaza at dusk, modern glass buildings, warm interior lights, cool ambient glow, evening sky, paved ground, distant street lamps."</blockquote><blockquote>"andrnml1, chest-up, straight-on, head tilted back, mouth open, laughing, pink background, vibrant blue and purple lighting, studio setting, even illumination, theatrical pose, expressive gesture, stylized appearance."</blockquote><blockquote>"andrnml1, three-quarter from mid-thigh up, straight-on, arms crossed, seated, stern, athletic track background, warm golden hour light, outdoor, bench, red lanes, green field, soft shadows."</blockquote><blockquote>"andrnml1, three-quarter from mid-thigh up, straight-on, hands on knees, stern expression, workshop with tools, bright overhead lighting, wooden workbenches, organized wall-mounted equipment, industrial ceiling fixtures."</blockquote><h3>Training</h3><ul><li>Base model: Krea 2 (trained on Raw)</li><li>Type: Character LoRA</li><li>Dataset: 33 images</li><li>Steps: 2000</li><li>Network rank / alpha: 32 / 32</li><li>Learning rate: 0.0002</li><li>Optimizer: adamw8bit</li><li>Resolutions: 512 / 1024</li></ul><h3>Generation settings</h3>Trained on Krea 2 Raw, and transfers to Krea 2 Turbo for faster generation.<ul><li>Raw: ~30 steps, guidance 3.5</li><li>Turbo: ~8 steps, guidance 1.0</li></ul>The preview images were generated with <code>krea2_turbo_mxfp8.safetensors</code>, at 8 steps, guidance 1, LoRA weight 1.

Files with SHA256: 48e19b3515c77473f2e200351621ce030b2539b6e1c2d992ad977ee32d41cbe5