Lora Model created from 50 public images of Blake Lively.
Warning: Might produce nudity if not appropiated prompt.
Supports are welcome : If you like my work, just buy me a beer (https://ko-fi.com/barabasj214).
Special requests: Do you want something in particular or a personal LoRA model? Contact me through my Ko-fi page sending a direct message (https://ko-fi.com/barabasj214).
Description
FAQ
Comments (3)
Hey, I'm running this lora through my usual tests (multi-model same-seed and lora-compatability) and I'm having some trouble easily replicating good results. For some clarification, can you let me know (and probably add to the description of this and future models): what model this was trained on and; were there any training words (I see you use BL in your prompts, but haven't listed it in trained words) and; what other models you have tested this on, with vae's if possible. I'm testing on about 3 realistic and 5 "semi" realistic models with the vae-ft-mse vae and I'm either getting what looks like an over trained image, or an undertrained, based on seed not model (as in the same seed is either over or under trained despite the model). If you want I can make and upload some images, but at the moment the one's I'm testing with would get taken down I think.
Hi! Thanks for your time reviewing my model. First, the model used for training this LoRA is "v1-5-pruned.safetensors". It is the same I always use. Have you tried Positive and Negative prompts exactly as I use them? Just to be sure it works Ok.
Hi Skittlz, the model used for training was "v1-5-pruned.safetensors". It is the same I always use. For your reference, not all of my results are perfect.