<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>Join the community on Discord: <a target="_blank" rel="ugc" href="https://discord.gg/nB79wRNtMT">https://discord.gg/nB79wRNtMT</a><h2>Triple Exposure</h2>A style LoRA for Krea 2 that turns a portrait into a triple exposure collage: the same person shown three times, as a large close-up portrait, a profile and a small full-body figure, all blended with city skylines, landmarks and landscapes. The edges dissolve into ink splashes and smoke over a light background, with a strong colour palette holding it together.It works with women and men, any age, look or outfit, and with any pair of places: Venice canals, Kyoto temples, the Manhattan skyline, a desert road, the Northern Lights…<h3>Usage</h3>Start with the trigger word <code>TripleExposureRK</code>. Describe the person, what they are wearing and their expression, then the collage, the two scenes it blends with and the colours:<code>TripleExposureRK, Photograph of a &lt;person&gt; with &lt;face&gt; and &lt;hair&gt;. She is wearing &lt;outfit&gt;, with &lt;expression&gt;. The image is a surreal triple exposure collage: a large close-up portrait, a profile and a small full-body figure of the same person, blended with &lt;scene 1&gt; and &lt;scene 2&gt;. The colors are &lt;palette&gt;, with ink splashes and smoke dissolving the edges.</code>Palettes that work well: black and white with red, gold or blue accents; warm sunset orange and magenta; soft pastel pink and teal; cool silver and icy blue; sepia with emerald green.Recommended weight: 1.0. Made for vertical images (9:16).<h3>Prompt Examples</h3><blockquote>"TripleExposureRK, Photograph of a 43-year-old Black woman with big doe eyes and a button nose and 1950s victory rolls in chestnut brown. She is wearing a corset top and a flowing skirt, with a confident pin-up smile. The image is a surreal triple exposure collage: a large close-up portrait, a profile and a small full-body figure of the same person, blended with Marrakech souks and lanterns and London's Big Ben and a red double-decker bus. The colors are vivid tropical turquoise and coral tones, with ink splashes and smoke dissolving the edges."</blockquote><blockquote>"TripleExposureRK, Photograph of a 45-year-old Nordic man with a round face and a warm smile and slicked-back black hair. He is wearing a navy peacoat and scarf, with a serious, determined expression. The image is a surreal triple exposure collage: a large close-up portrait, a profile and a small full-body figure of the same person, blended with Seoul's palaces and modern towers and the temples of Angkor in the jungle. The colors are warm sunset orange and magenta tones, with ink splashes and smoke dissolving the edges."</blockquote><blockquote>"TripleExposureRK, Photograph of a 32-year-old Latina woman with almond eyes and a sharp jawline and long straight jet-black hair. She is wearing a vintage aviator jacket and scarf, with a sultry look over her shoulder. The image is a surreal triple exposure collage: a large close-up portrait, a profile and a small full-body figure of the same person, blended with Kyoto temples under cherry blossoms and the Hong Kong harbour skyline. The colors are cool silver and icy blue tones, with ink splashes and smoke dissolving the edges."</blockquote><blockquote>"TripleExposureRK, Photograph of a 24-year-old Middle Eastern woman with big doe eyes and a button nose and box braids tied up high. She is wearing a silk kimono-style robe, with a playful wink. The image is a surreal triple exposure collage: a large close-up portrait, a profile and a small full-body figure of the same person, blended with Istanbul's mosques over the Bosphorus and the canals and bridges of Venice. The colors are black and white with crimson and gold accents, with ink splashes and smoke dissolving the edges."</blockquote><h3>Training</h3><ul><li>Base model: Krea 2</li><li>Type: Style LoRA</li><li>Steps: 4000</li><li>Rank: 32</li><li>Resolution: 1024</li><li>Optimizer: adamw8bit, learning rate 0.0001</li><li>Engine: ai-toolkit</li></ul><h3>Generation settings</h3>Trained on Krea 2, and works on 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 gallery images were generated with <code>krea2_turbo_mxfp8.safetensors</code>, at 8 steps, guidance 1, LoRA weight 1, 768×1368.

Files with SHA256: 7119eb31f2c7833e1fca51d2640f24231444e2f350aece1ec5adb13efedddedc