<h3 id="anima-v1">Anima V1</h3>Trained base on Anima V0.1 base with LoRA_Easy_Training_Scripts. See about this version for training details.<h3 id="chenkin-rectified-flow-v1">Chenkin Rectified-Flow V1</h3>Trained base on Chenkin Rectified-Flow 0.3 with LoRA_Easy_Training_Scripts. See about this version for training details.<h3 id="noob-v-pred-v6">NOOB V-PRED V6</h3>Retrained with updated dataset. The images are brighter. But the skin texture downgrades a little.<h3 id="noob-v-pred-v5">NOOB V-PRED V5</h3>Retrained with updated dataset and parameters. Skin texture and bodyshape improved.<h3 id="noob-v-pred-v4">NOOB V-PRED V4</h3>Updated dataset to prevent "arona" influence and to achieve better detail. Way too many arona pictures in the last version dataset. Skin texture also improved.<h3 id="noob-v-pred-v3">NOOB V-PRED V3</h3>Updated dataset. Add dark theme in prompt to achieve better results.<h3 id="noob-v-pred-v2">NOOB V-PRED V2</h3>Updated dataset.<h3 id="introduction">Introduction</h3>A Lycoris to generate the AI artist nyalia like style. Trained based on NOOB V-PRED 1.0 with OneTrainer. Dataset included works of nyalia and similar artists.<h3 id="training-details">Training details</h3>See about the version.

Nyalia like style

[[subsets]]name = "5"image_dir = "G:/dataset/style/nyalialike_V6/5_Cunnyfunnky"num_repeats = 5shuffle_caption = truecaption_extension = ".txt"random_crop_padding_percent = 0.05caption_dropout_rate = 0.1caption_tag_dropout_rate = 0.1[[subsets]]name = "5_1780505591313840500"image_dir = "G:/dataset/style/nyalialike_V6/5_kampa"num_repeats = 5shuffle_caption = truecaption_extension = ".txt"random_crop_padding_percent = 0.05caption_dropout_rate = 0.1caption_tag_dropout_rate = 0.1[[subsets]]name = "5_1780505591328839600"image_dir = "G:/dataset/style/nyalialike_V6/5_lovelace-12"num_repeats = 5shuffle_caption = truecaption_extension = ".txt"random_crop_padding_percent = 0.05caption_dropout_rate = 0.1caption_tag_dropout_rate = 0.1[[subsets]]name = "5_1780505591345356700"image_dir = "G:/dataset/style/nyalialike_V6/5_NFFA"num_repeats = 5shuffle_caption = truecaption_extension = ".txt"random_crop_padding_percent = 0.05caption_dropout_rate = 0.1caption_tag_dropout_rate = 0.1[[subsets]]name = "5_1780505591361355200"image_dir = "G:/dataset/style/nyalialike_V6/5_Nyako"num_repeats = 5shuffle_caption = truecaption_extension = ".txt"random_crop_padding_percent = 0.05caption_dropout_rate = 0.1caption_tag_dropout_rate = 0.1[[subsets]]name = "5_1780505591377864800"image_dir = "G:/dataset/style/nyalialike_V6/5_Nyalia"num_repeats = 5shuffle_caption = truecaption_extension = ".txt"random_crop_padding_percent = 0.05caption_dropout_rate = 0.1caption_tag_dropout_rate = 0.1[train_mode]train_mode = "lora"[general_args.args]persistent_data_loader_workers = truevae_batch_size = 5pretrained_model_name_or_path = ""mixed_precision = "bf16"gradient_checkpointing = truegradient_accumulation_steps = 1seed = 42max_data_loader_n_workers = 1max_token_length = 225prior_loss_weight = 1.0sdpa = truemax_train_epochs = 25cache_latents = truecache_latents_to_disk = true[general_args.dataset_args]resolution = 1024batch_size = 1[network_args.args]network_dim = 64network_alpha = 1.0min_timestep = 0max_timestep = 1000network_train_unet_only = true[optimizer_args.args]optimizer_type = "ProdigyPlusScheduleFree"lr_scheduler = "constant"loss_type = "l2"learning_rate = 1.0unet_lr = 1.0max_grad_norm = 1.0min_snr_gamma = 1.0[saving_args.args]output_dir = "G:/LoRA_Easy_Training_Scripts/output"output_name = "nyalialike-anima1.0-V1-locon-dim64conv16alpha0.01-SNR1"save_precision = "bf16"save_model_as = "safetensors"save_every_n_epochs = 1save_toml = truesave_toml_location = "G:/LoRA_Easy_Training_Scripts/output"[sample_args.args]sample_sampler = "euler"sample_every_n_epochs = 1sample_prompts = "G:/LoRA_Easy_Training_Scripts/anima sample.txt"[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_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 = 512t5_max_token_length = 512timestep_sampling = "sigmoid"sigmoid_scale = 1.0discrete_flow_shift = 3.0[edm_loss_args.args]edm2_loss_weighting = false[bucket_args.dataset_args]enable_bucket = truemin_bucket_reso = 256max_bucket_reso = 2048bucket_reso_steps = 64[network_<a target="_blank" rel="ugc" href="http://args.args.network">args.args.network</a>_args]conv_dim = 16conv_alpha = 1.0train_llm_adapter = "False"[optimizer_args.args.optimizer_args]betas = "0.9,0.99"beta3 = "None"weight_decay = "0"weight_decay_by_lr = "True"d0 = "1e-6"d_coef = "2"d_limiter = "True"prodigy_steps = "0"schedulefree_c = "0"eps = "1e-8"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"

Files with SHA256: 723aaa27543feff07bef9d6b46aae5a2383f44905a887fe8776a1d3e8e9ed686