This is for all the freaks out there. The aim of this lora is to help you generate / detail humanoid monster peens of different shapes and sizes.
You'll see as part of this package there is a main "V2 Base" lora and a "V2 Detailer" lora. TLDR, use the base lora to generate your images. 0.8-1.0 strength is the sweet spot but it heavily depends on your checkpoint. I find Indigo works well. The detailer lora is there to be plugged into your adetailer or fdetailer to enhance the penis details. It can also be used on images that were generated without the base lora as seen in the detailer's sample pics. I go into more detail on how they work in my canine penis lora if you're interested.
The tags are on the right for all the different types of peens. Feel free to mix and match. Some work better than other.
For more info on how to use these loras. Here's a quick how to:
Generate your base, undetailed image the same way you usually would. Either add the base lora to this step or don't. If you do, add "m0nster hum4noid pp" as a tag as well as any of the other tags on the right. You can stop right here, but if you want to go further...
Plug the detailer lora into either your adetailer or fdetailer.
Add the tag "m0nster hum4noid pp" to your detailer prompt if it isn't in your base prompt. As well as the other tags you want.
Adjust the weight of the lora and the denoise of the detailer as you want. In my testing, a weight of 1 on the lora and a denoise of 0.40 is the sweet spot. The higher the denoise the higher the risk/reward.
If you're struggling, check the quick example workflows I have attached to the sample pics.
Description
Retagged and retrained. Now includes 100 hand-created context images to better allow the lora to be used without affecting the overall image style. Its a direct upgrade over V1 imo.
I also wanted to be more "honest" with the sample pics this time. There's much less added detailer weights etc to show the model's capabilities to those without super complex Comfyui flows.
Trained on 433 images:
{
"engine": "kohya",
"unetLR": 0.0005,
"clipSkip": 2,
"loraType": "lora",
"keepTokens": 1,
"networkDim": 16,
"numRepeats": 3,
"resolution": 1024,
"lrScheduler": "cosine",
"minSnrGamma": 5,
"noiseOffset": 0.1,
"targetSteps": 6495,
"enableBucket": true,
"networkAlpha": 8,
"optimizerType": "Prodigy",
"textEncoderLR": 0.00005,
"maxTrainEpochs": 20,
"shuffleCaption": true,
"trainBatchSize": 4,
"flipAugmentation": true,
"lrSchedulerNumCycles": 3
}
FAQ
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.