penis, small_penis, big_penis, pubes, trimmed_pubes, shaved_pubes, black_hair, brown_hair, blond_hair, ginger_hair, flaccid, saggy
๐ Version 4.3 is here with a major update.V4.3E21 can generate realistic images with 5 times fidelity than V4.2 theoretically. New tagging method allows better control on penis size, pubes and balls. See below for full update note, use instructions and checkpoint recommendations.๐ Important instructions:V4.3E21 (5 times fidelity): When working with other Loras, start with a lora weight of 0.7 and a low CFG scale 5.5 (a higher CFG scale results in sharper outputs).V4.3E41 (10 times fidelity): This ultra-fidelity version only works as a standalone. The output quality surpassed that of images from flux.๐ Checkpoint recommendations: ๐ Share your pics for training a better lora[NOTE] 1) untrimmed 2) healthy (no varix, no loose urethra, no pearly penile papules), hydrated, natural state from various angles (take a cold shower helps) 3) natural light 4) no bokeh / no foreshortening / no hand / no clothes / no overweight / no face / day light / simple background / no close up / 5) below navel, above knees 6) don't use front camera / use RAW mode (as artificial fusion and smartHDR from iPhone are poisons to training files)๐ Two uses of this lora, create real life penis with triggering word "penis", anime style dominated checkpoint requires additional triggering word "realistic" .๐ Enhance realness of a photo, especially on hairs without using triggering word.Use the Lora with epoch below 7 for a stable result, like V03E06, though lora with higher epoch like V03E10 has better resemblances to the training source; (this does not apply to V4 or later๏ผUSE fewer triggering word each time for best result;Use CFG between 4.5 (complex scene) to 7.5 (simple);For anime start training steps from a lower value like 19 with Euler a sampler, if you are satisfied with the result, use hi-res fix / face fix (and set Euler to 29) to increase over all quality;For realistic photos, use DPM++2M Karra, 30+ steps for close up portrait, 35+ steps for full body shot.There's no universal prompt / lora weight for every generation. If you're using anime style / character lora with realistic checkpoint, starting with a lower lora weight like 0.7. Ignore the lora weight in sample images, due to a recent bug, image generated online will display lora weight of 1 regardless the actual value.Step number setting (which is the weight for all tags) is closely associated with the complexity of the prompt, if your outcome has extra objects like fingers or limbs, try lower the step value (and the lora weight too); if your outcome has objects disappeared, lack of quality, texture and reality, try raising the step value.๐ Changes in Version 4.3:1) Use tag with underscore.2) Trained with "slow baking" technique with proper num repeat and epoch ratio.๐ Changes in Version 4.2:1) New racial ratio: 7Afr | 3Lat | 87Cau | 71Asi2) New state ratio: 60Flaccid | 38Erect | 66Semi-erect๐ Changes in Version 4.1:1) Some of the source photos from V3 include limbs that needed to be cropped out or removed. They make V3 loose focus on penis, creating distortion and diphallia which never happened on V2.2) Remove resources with weird foreskin3) Reduce training resource quantity down to ???, as this is a common problem across most of my loras, too many training images reduce the outcome quality, the quality of the resource images is more important.4) Increase training resource resolution.5) Reduce training steps down to 1000, as I found lower training epochs lora got stable results.
