Ver Anima V1
Trained base on Anima V0.1 base with LoRA_Easy_Training_Scripts. See about this version for training details.
Ver Noob V-PRED V3
Trained on 60 images generated with V2. An equal number of 1girl regularization set was added to enhance art style and reduce overfitting. The base model is still NOOB V-PRED V1.0.
In this version, the stability of vein, prompt responsiveness and anatomy were better. It improved adaptability in different models also. The generated veins are a bit too large, which is my personal preference. But if you feel too large for you, you can generate smaller ones by reducing the weight or removing the keyword: veiny penis.
Ver Noob V-PRED V2
Merged with several loras trained based on NOOB V-PRED V1.0. Removed some blocks from realisitc loras to reduce artstyle influence and achieve better results.
Pros:
Better veins
Cons:
Limb distortion
Ver Noob V-PRED V1
Merged with several loras trained based on NOOB V-PRED V0.65S.
Pros:
Better shape and texture of penis
Better skin texture
Cons:
Style influence
Limb distortion
Ver Noob Exprimental
Trained based on NOOB E-PRED V1.0. See about the version for training details.
When applying this lora, your generation will have darker body and thicker line compared to original style you use. This is an issue of dataset style influence. You can lower the weight of lora, trigger word or CFG. Or you may upweight the style words or lora to reduce the style influence.
Introduction
A Pony based LyCORIS to enlarge penis veins.
Pros
Concept generation
Cons
Penis seems to be larger. But "small penis" would work.
Training details
See about the version.
Description
[[subsets]]
caption_dropout_rate = 0.05
caption_extension = ".txt"
caption_tag_dropout_rate = 0.05
image_dir = "G:/dataset/concept/veiny_V1/5_veinypp"
keep_tokens = 1
name = "20"
num_repeats = 5
random_crop_padding_percent = 0.05
shuffle_caption = true
[train_mode]
train_mode = "lora"
[general_args.args]
persistent_data_loader_workers = true
vae_batch_size = 5
pretrained_model_name_or_path = ""
mixed_precision = "bf16"
gradient_checkpointing = true
gradient_accumulation_steps = 1
seed = 42
max_data_loader_n_workers = 1
max_token_length = 225
prior_loss_weight = 1.0
sdpa = true
max_train_epochs = 20
cache_latents = true
cache_latents_to_disk = true
[general_args.dataset_args]
resolution = 1024
batch_size = 5
[network_args.args]
network_dim = 32
network_alpha = 16.0
min_timestep = 0
max_timestep = 1000
network_train_unet_only = true
[optimizer_args.args]
optimizer_type = "ProdigyPlusScheduleFree"
lr_scheduler = "constant"
loss_type = "l2"
learning_rate = 1.0
unet_lr = 1.0
max_grad_norm = 1.0
[saving_args.args]
output_dir = "G:/LoRA_Easy_Training_Scripts/output"
output_name = "veiny-anima1.0-V1-loha-dim32conv16AR0.5-NewTag"
save_precision = "bf16"
save_model_as = "safetensors"
save_every_n_epochs = 1
save_toml = true
save_toml_location = "G:/LoRA_Easy_Training_Scripts/output"
[logging_args.args]
log_prefix_mode = "disabled"
run_name_mode = "default"
[anima_args.args]
pretrained_model_name_or_path = "G:/sd-webui-forge-neo/models/Stable-diffusion/ANIMA/anima_baseV10.safetensors"
qwen3 = "G:/sd-webui-forge-neo/models/text_encoder/qwen_3_06b_base.safetensors"
vae = "G:/sd-webui-forge-neo/models/VAE/qwen_image_vae.safetensors"
qwen3_max_token_length = 512
t5_max_token_length = 512
timestep_sampling = "sigmoid"
sigmoid_scale = 1.0
[edm_loss_args.args]
edm2_loss_weighting = false
[bucket_args.dataset_args]
enable_bucket = true
min_bucket_reso = 256
max_bucket_reso = 2048
bucket_reso_steps = 64
[network_args.args.network_args]
conv_dim = 16
conv_alpha = 8.0
algo = "loha"
train_llm_adapter = "false"
[optimizer_args.args.optimizer_args]
betas = "0.95,0.99"
beta3 = "None"
weight_decay = "0"
weight_decay_by_lr = "True"
d0 = "2e-5"
d_coef = "1"
d_limiter = "True"
schedulefree_c = "0"
eps = "None"
split_groups = "True"
split_groups_mean = "False"
factored = "True"
factored_fp32 = "True"
use_bias_correction = "True"
use_stableadamw = "True"
use_schedulefree = "True"
use_speed = "False"
stochastic_rounding = "True"
fused_back_pass = "False"
use_cautious = "False"
use_grams = "False"
use_adopt = "False"
use_orthograd = "False"
use_focus = "False"