<h1 id="overview">Overview</h1>A clothed version of my <a target="_blank" rel="ugc" href="https://civarchive.com/models/2294035">other</a> macromastia LoRA. This one here cannot produce X or XXX reliably.The model is pretty straight forward. Use the trigger early in your prompt. Further descriptions to the breasts position, framing and focus can help a lot to emphasize the effect.Sometimes it creates thicker or chubby bodies, for that you can include descriptors like <code>slim</code> or <code>skinny</code> or use negative prompts.Values over <code>1.0</code> may produce heavy artifacts. Decrease strength if you encounter artifacts. Most samples use a strength of <code>0.9</code>.For more, have a look at the example images/videos.<h3 id="ltxv-2.3-(sulphur-2)-v0">LTXV 2.3 (Sulphur 2) - V0</h3>Coming in June...<h2 id="ideogram-4-v1">Ideogram 4 - V1</h2>Experimental first draft. Things break and body horror may occur.Trained on bounding boxes and JSON. Use the trigger <code>gigantic breasts</code> on a bounding box that should represent the woman's breasts. This way, the size and position can sort of be controlled.<ul><li>Training model: <a target="_blank" rel="ugc" href="https://huggingface.co/ideogram-ai/ideogram-4-fp8">Ideogram 4 fp8</a></li><li>LoRA strength: 0.5-1.0</li><li>Sampler: euler / euler a</li><li>Steps: 20+</li></ul><pre><code>{...} // have a look at the showcase prompts</code></pre><h3 id="wan-2.2-t2v-v1">WAN 2.2 T2V - V1</h3>After using the trigger <code>macromastia</code>, use additional triggers <code>large</code>, <code>huge</code> and <code>gigantic</code> to control the size further, see example video. Can be combined with <a target="_blank" rel="ugc" href="https://civarchive.com/models/2322189/thiccification-formerly-body-expansion">Thiccification</a>.<ul><li>Training model: <a target="_blank" rel="ugc" href="https://civarchive.com/models/1852647?modelVersionId=2096603">WAN 2.2 T2V</a></li><li>LoRA strength conditional: 0.7-1.0</li><li>LoRA strength unconditional: 0.3-0.5</li><li>Sampler: euler / euler a</li><li>Use <a target="_blank" rel="ugc" href="https://github.com/VraethrDalkr/ComfyUI-TripleKSampler">TripleKSampler</a> with 1/4/4 steps or no lightning LoRAs at all for quality.</li><li>Use breast physics LoRAs to improve jiggle and motion for very large breasts. I highly recommend <a target="_blank" rel="ugc" href="https://civarchive.com/models/1852647/m4crom4sti4-huge-natural-breasts-physics-wan22-video-lora-k3nk">Huge Natural Breasts Physics</a> at strength 0.3-0.5 for the high noise, no low noise needed or it will degrade texture quality.</li></ul><pre><code>macromastia, a 21-year-old ginger woman drinking a milkshake from a straw in a cafe. she's wearing a tight blouse and jeans. She has gigantic breasts and a large cleavage. Her face is pretty and freckled, she has big lips. Looking out the window, sitting at a table.</code></pre><h3 id="flux.2-klein-9b-v2">Flux.2 [klein] 9B - V2</h3>Use the trigger <code>mcrmst</code> and describe the breast/cleavage in some detail.<ul><li>Training model: <a target="_blank" rel="ugc" href="https://huggingface.co/black-forest-labs/FLUX.2-klein-base-9B">Flux.2 [klein] Base 9B</a></li><li>LoRA strength: 0.3-1.0</li><li>Sampler: euler / res_2s</li><li>Steps: 4+</li><li>CFG: 1.0</li></ul><pre><code>mcrmst, A photograph of a woman with gigantic breasts, her voluminous, heavy breasts dramatically stretching and filling the soft beige cable-knit turtleneck sweater, creating deep curves and visible fabric tension despite the modest high neckline, sitting on a picnic blanket in a serene autumn forest.</code></pre><h3 id="zib-v1">ZIB - V1</h3>This model may also work with Z Image Turbo on higher strengths (1.0-2.0). You can also give <a target="_blank" rel="ugc" href="https://huggingface.co/GuangyuanSD/Z-Image-Distilled">Z Image Distilled</a> a try.After using the trigger <code>macromastia</code>, use additional triggers <code>large</code>, <code>huge</code> and <code>gigantic</code> to control the size further.Experimenting with negative prompts and strengths can be very rewarding.<ul><li>Training model: <a target="_blank" rel="ugc" href="https://huggingface.co/Tongyi-MAI/Z-Image">Z Image</a></li><li>LoRA strength: 0.3-1.0 (up to 2.0 on Turbo)</li><li>Sampler: res_2s+beta57 / euler / res_multistep / DPM++ 3M SDE</li><li>Steps: 40+ (8+ on Turbo)</li><li>CFG: 3.0-5.0 (1.0 on Turbo)</li></ul><pre><code>macromastia, A high quality photo of a beautiful young woman with long flowing auburn hair standing gracefully as the central figure in a vast field of tall sunflowers during golden hour sunset. She wears a light, flowing white summer dress that catches the warm breeze and is stretched over her large breasts, her skin glowing in soft natural light, subtle smile on her face.</code></pre><h3 id="zit-v2">ZIT - V2</h3>Not compatible with Z Image Base.<ul><li>Training model: <a target="_blank" rel="ugc" href="https://huggingface.co/ostris/Z-Image-De-Turbo">Z Image De-Turbo</a></li><li>LoRA strength: 0.3-1.0</li><li>Sampler: euler / res_multistep / DPM++ 3M SDE</li><li>Steps: 8+</li><li>CFG: 1.0</li></ul>Prompt example:<pre><code>A photo of a cute woman with macromastia lying on a beach towel, wearing a yellow bikini revealing her breasts. Her breasts spill out the top of her bikini, forming a significant cleavage. She's bathing in the sun smiling at the camera. A summer day and warm candid setting.</code></pre>

🍒 Macromastia (Clothed) [Ideogram 4, Flux, Z-Image, etc.]

More data and more steps. Should perform slightly better with more variation. Lying position still kinda undertrained and require more careful prompting.

Files with SHA256: 2b6e10f865d8699240a54f4810753a1def28b2c775e0c833578404cb7acb24c4