Ver NOOB V-PRED V3 EQ
Trained with improved parameters based on NOOB V-PRED EQ VAE exprimental version: https://civarchive.com/models/1858821/noobai-v-pred-10-with-eq-vae
In this version, I subdivided the trigger words for more accurate generate. 10 triggerwords set. See the examples first.
NOOB V-PRED V2
Trained base on NOOB V-PRED 1.0 with Kohya SS. See about this version for training details. Dataset updated and triggerwords changed.
The main upgrade of V2 is the understanding of multi futa scenes after changing training parameters and dataset, as well as the compatibility of finetuned checkpoints and multi loras. But it still need some trials to get the desired generation.
NOOB V-PRED V1
Trained base on NOOB V-PRED 1.0 with OneTrainer. See about this version for training details. Dataset updated and triggerwords changed.
The reason for this update is when I applied multiple Loras at the same time, increasing the weight of gigantic penis to 2 still couldn't achieve the result I needed. I think this Lora will solve the problem of giant pp generation when multiple Loras work together. After testing, the effect is good.
About PonyXL V2.0
Used regularization techniques and updated dataset to get less influence on artstyle and better penis shape. Also improved the effect when using other loras together. I chose Lora with less bake to achieve better interaction, but the disadvantage is that in some scenarios, the weight of penis needs to be increased.
Added "horse penis" for penis type.
About PonyXL
Used PonyXL to train this one. Updated dataset to achieve better performance. It enhanced this concept on PonyXL and related models.
Adding a concept: gigantic_dark
About V3.0
Updated dataset and used multires technique in this version. It would bring more light and shadow with quality tag, such as masterpiece, highly_detailed, best_quality ,etc.
Bewared that may affect the result of generation. You may get chaos result if you use these tags and overweight them.
Multires technique parameter:
multires_noise_iterations="6"
multires_noise_discount=0.3
IMPORTANT: Now I set the trigger words as gigantic_penis and gigantic_penis_ejaculation.
About V2.0
In v2.0, I removed all the images with black bar censor and retrain this locon. That will do help to get better result. The annoying censor bar won't be there.
I set batch size = 2 to acclerate training. The final results are good.
Introduction
This LyCORIS is used to generate futanari with gigantic penis. If you put it in a long picuture like (1024X768), you can get a massive long cock. You may find the faces are blurry but you can fix it with inpaint or img to img. Adding hyper futanari LyCORIS can improve the texture of penis.
How to use
I recommand using it in weight 0.75-0.85. When working with other loras, it is not easy to control. But you can get what you want after some tries.
You may get good result in lower weight when use it alone. Stiil, it needs more tests.
Training details:
Traingset is of about 145 images. They are mirrored before training. Half of them are chosen as regularization images. Total steps are about 8800.
Regularization = true
resolution=512
batch_size=2
epoch=10
network_dim=32
network_alpha=32
clip_skip=2
Using AdamW8bits as optimizer:
lr="1e-4"
unet_lr="1e-4"
text_encoder_lr="1e-5"
Locon parameters:
conv_dim=4
conv_alpha=4
Description
{
"LoRA_type": "LyCORIS/LoCon",
"LyCORIS_preset": "full",
"adaptive_noise_scale": 0,
"additional_parameters": "",
"async_upload": false,
"block_alphas": "",
"block_dims": "",
"block_lr_zero_threshold": "",
"bucket_no_upscale": true,
"bucket_reso_steps": 128,
"bypass_mode": false,
"cache_latents": false,
"cache_latents_to_disk": false,
"caption_dropout_every_n_epochs": 0,
"caption_dropout_rate": 0,
"caption_extension": ".txt",
"clip_skip": 2,
"color_aug": false,
"constrain": 0,
"conv_alpha": 4,
"conv_block_alphas": "",
"conv_block_dims": "",
"conv_dim": 8,
"dataset_config": "",
"debiased_estimation_loss": false,
"decompose_both": false,
"dim_from_weights": false,
"dora_wd": false,
"down_lr_weight": "",
"dynamo_backend": "no",
"dynamo_mode": "default",
"dynamo_use_dynamic": false,
"dynamo_use_fullgraph": false,
"enable_bucket": true,
"epoch": 10,
"extra_accelerate_launch_args": "",
"factor": -1,
"flip_aug": false,
"fp8_base": false,
"full_bf16": false,
"full_fp16": false,
"gpu_ids": "",
"gradient_accumulation_steps": 1,
"gradient_checkpointing": true,
"huber_c": 0.1,
"huber_schedule": "snr",
"huggingface_path_in_repo": "",
"huggingface_repo_id": "",
"huggingface_repo_type": "",
"huggingface_repo_visibility": "",
"huggingface_token": "",
"ip_noise_gamma": 0.1,
"ip_noise_gamma_random_strength": false,
"keep_tokens": 1,
"learning_rate": 1,
"log_tracker_config": "",
"log_tracker_name": "",
"log_with": "tensorboard",
"logging_dir": "/root/autodl-tmp/output",
"loss_type": "l2",
"lr_scheduler": "cosine",
"lr_scheduler_args": "",
"lr_scheduler_num_cycles": 1,
"lr_scheduler_power": 1,
"lr_warmup": 0,
"main_process_port": 0,
"masked_loss": false,
"max_bucket_reso": 4096,
"max_data_loader_n_workers": 0,
"max_grad_norm": 1,
"max_resolution": "1024,1024",
"max_timestep": 1000,
"max_token_length": 75,
"max_train_epochs": 0,
"max_train_steps": 0,
"mem_eff_attn": false,
"metadata_author": "",
"metadata_description": "",
"metadata_license": "",
"metadata_tags": "",
"metadata_title": "",
"mid_lr_weight": "",
"min_bucket_reso": 1024,
"min_snr_gamma": 0,
"min_timestep": 0,
"mixed_precision": "bf16",
"model_list": "custom",
"module_dropout": 0,
"multi_gpu": false,
"multires_noise_discount": 0.1,
"multires_noise_iterations": 6,
"network_alpha": 8,
"network_dim": 16,
"network_dropout": 0,
"network_weights": "",
"noise_offset": 0,
"noise_offset_random_strength": false,
"noise_offset_type": "Multires",
"num_cpu_threads_per_process": 2,
"num_machines": 1,
"num_processes": 1,
"optimizer": "Prodigy",
"optimizer_args": "\"weight_decay=0.01\" \"d_coef=1\" \"use_bias_correction=True\" \"safeguard_warmup=False\" \"betas=0.9,0.99\" \"decouple=True\"",
"output_dir": "/root/autodl-tmp/output",
"output_name": "Gigantic Futa V2-PonyXL",
"persistent_data_loader_workers": false,
"pretrained_model_name_or_path": "/root/kohya_ss/models/ponyDiffusionV6XL.safetensors",
"prior_loss_weight": 1,
"random_crop": false,
"rank_dropout": 0,
"rank_dropout_scale": false,
"reg_data_dir": "/root/autodl-tmp/reg_while_penetrating",
"rescaled": false,
"resume": "",
"resume_from_huggingface": "",
"sample_every_n_epochs": 1,
"sample_every_n_steps": 0,
"sample_prompts": "score_9, score_8_up, score_7_up, anime BREAK, gigantic_cock, 1girl, long hair, breasts, looking at viewer, open mouth, 1boy, red eyes, navel, jewelry, nipples, purple hair, hetero, outdoors, earrings, nude, sky, tongue, pointy ears, solo focus, day, cloud, tongue out, spread legs, mole, huge breasts, blue sky, completely nude, uncensored, saliva, erection, squatting, elf, mole under mouth, puffy nipples, veins, mole on breast, veiny penis, large areolae, huge nipples, animal penis, fellatio gesture, horse penis, gigantic penis --n source_pony, source_furry, source_cartoon, greyscale, from below --w 1024 --h 1024\n ",
"sample_sampler": "euler_a",
"save_every_n_epochs": 1,
"save_every_n_steps": 0,
"save_last_n_steps": 0,
"save_last_n_steps_state": 0,
"save_model_as": "safetensors",
"save_precision": "bf16",
"save_state": false,
"save_state_on_train_end": false,
"save_state_to_huggingface": false,
"scale_v_pred_loss_like_noise_pred": false,
"scale_weight_norms": 0,
"sdxl": true,
"sdxl_cache_text_encoder_outputs": false,
"sdxl_no_half_vae": true,
"seed": 31337,
"shuffle_caption": true,
"stop_text_encoder_training_pct": 0,
"text_encoder_lr": 1,
"train_batch_size": 5,
"train_data_dir": "/root/autodl-tmp/gigantic_futa_v2",
"train_norm": false,
"train_on_input": true,
"training_comment": "",
"unet_lr": 1,
"unit": 1,
"up_lr_weight": "",
"use_cp": false,
"use_scalar": false,
"use_tucker": false,
"v2": false,
"v_parameterization": false,
"v_pred_like_loss": 0,
"vae": "/root/kohya_ss/models/sdxl_vae_fp16fix.safetensors",
"vae_batch_size": 0,
"wandb_api_key": "",
"wandb_run_name": "",
"weighted_captions": false,
"xformers": "xformers"
}FAQ
Comments (7)
in your pony v2 description, you put "pennis" instead of penis, just letting you know XD
thx
@DeepDark_Fantasy514 no probs :D
Very nice. However, I am using ComfyUI. And the PonyXL version 2.0 does not work (here) at all. v1.0 works fine though. Just fyi.
May I ask why it's very easy to generate solo futa by this model,I'M trying to generate 1futa with 1girl version.
Training dataset used only consists of solo 1futa. So you need to upweight 2girls or futa with girl, or downweight the lora.
use regional prompting and use the lora only in the region with the futa.