Generated a round-face girl.
This version of LoRA was trained on three different Pony-based checkpoints.
mq_pony_64_64: Pony Diffusion V6 XL
mq_realP_64_64: real pony
mq_iniverse_64_64: iNiverse Mix XL(SFW & NSFW)
Training details are shown below.
Network Rank (Dimension): 64
Network Alpha: 64
Training data: 11 pictures with 1024×1024 resolution
max_train_steps: 1600
Number of Repeats: 30
The goal is to generate an Asian woman with a round face.
Suggested sampler: Euler a
I have tried different versions of LoRA with various checkpoints。
Prompt: score_9, score_8_up, score_7_up, score_6_up,score_anime,1girl,medium breasts, china dress, <lora:mq_pony_64_64:1> , zPDXL2,mq,close-up,detailed face,upper body, sfw,looking at viewer, long hair
Negative Prompt: worst quality, low quality, normal quality, lowres, bad anatomy, bad hands, source_pony, comic, 3d,middle age,adult,mother,mature female, milf,Muscles,censor, bar censor, mosaic censorship, zPDXL2-neg
Euler a, CFG Scale: 7, Seed: 191356349, 1024x1024, Steps: 25, Clip Skip: 2.
For Pony Diffusion V6 XL with different LoRA:
For real pony with different LoRA::
For iNiverse Mix XL(SFW & NSFW) with different LoRA:
Description
Trained on Pony Diffusion V6 XL.