💡 Search "huggy" to find all my models on site.💡 Models are updated frequently, so be sure to check back and use the latest version.💡 DO NOT MIXED USE Illustrious lora with SDXL / PONY / FLUX model.📒 This lora does:<ul><li>Bring "zoom blur", "chromatic aberration" and "black-and-white" effects to your works.</li><li>Create better male genitals.</li></ul>📒 Usage Tips:<ul><li>Works magically well with "Plant Milk 🌿 - Model Suite".</li><li>Set lora strength 0.8 or lower</li><li>For realistic styles, use DPM++ 2M Karras sampler</li><li>For expressive anime-style eyes, Euler tends to perform better</li><li>When using with other lora, set the other lora weights to 0.8 or lower</li><li>If Illustrious models produce stubborn 4-fingered hands, use a passive prompt like (5 fingers:1.7)</li><li>The Illustrious lora is not compatible with SDXL or Pony models</li></ul>📒 Essential trigger words:<ul><li>chromatic aberration</li></ul>📒 Optional trigger words:<ul><li>penis / small penis / saggy testicles / hairy / pubes</li></ul>📒 Changes in V01E81:<ul><li>Image count increased from 82 to 94</li><li>Repeats per image increased from 2 to 3</li><li>UNet learning rate raised from 0.0003 to 0.0004</li><li>Min SNR Gamma increased from 3 to 4</li></ul><ul><li>V01E81 was trained over 7,200 steps, this version delivers approximately 1.6× stronger chromatic aberration effects and improved generalization compared to previous versions.</li></ul><ul><li>V00E92 the initial version was trained using very slow "baking" settings recommended by Gemini. The result is noticeably underfitted, with very subtle effects.</li></ul>📒 Cost Breakdown for This Lora:1,100 Buzz – 90 high-resolution source file generations3,600 Buzz – Training 2 rounds3,900 Buzz – TestingTotal: 9,600 Buzz🎉 This lora is part of my goal to create 100 lora.📒 MirroringI’ve noticed a few sites, like Modelslab and yodayo, using bots to mirror my models. SeaArt in particular shows big numbers for downloads and generations.<ol><li>I don’t mind if you mirror my models — sites like civitaiarchive have always done that.</li><li>I do mind if you block users from downloading the models and force them onto a paid online service.</li><li>I do mind if you claim copyright or give users ownership of the generations. These are open-source models, trained on open-source bases with limited-ownership data. No one — not even me — should claim copyright. Credit is all that’s needed.</li></ol>

Prism 🌈 by Huggy

{"Image": "82","engine": "kohya","unetLR": 0.0003,"clipSkip": 1,"loraType": "lora","keepTokens": 1,"networkDim": 32,"numRepeats": 2,"resolution": 1024,"lrScheduler": "cosine_with_restarts","minSnrGamma": 3,"noiseOffset": 0.08,"targetSteps": 3690,"enableBucket": true,"networkAlpha": 16,"optimizerType": "Adafactor","textEncoderLR": 0.00005,"maxTrainEpochs": 90,"shuffleCaption": true,"trainBatchSize": 4,"flipAugmentation": true,"lrSchedulerNumCycles": 3}

Files with SHA256: d56af3da8f4b881b428942585c6ff46ca4951a122a5e5b079717b3cec272ad49