"It's me, your fiancée, Ruto! Princess of the Zoras! I never forgot the vows we made to each other seven years ago!" — Princess Ruto
Follow me on X/Twitter for exclusive content: https://x.com/Citron_Legacy 😁
Please checkout my collection of over 50 Legend of Zelda Loras: https://civarchive.com/collections/266
Description
Trained on: 200 images
Repeats: 1
Steps: 440
Total Time: 1:19:25
--
[network_arguments]
unet_lr = 0.75
text_encoder_lr = 0.75
network_dim = 32
network_alpha = 32
network_module = "networks.lora"
network_train_unet_only = false
[optimizer_arguments]
learning_rate = 0.75
lr_scheduler = "cosine_with_restarts"
lr_scheduler_num_cycles = 3
lr_warmup_steps = 20
optimizer_type = "Prodigy"
optimizer_args = [ "weight_decay=0.1", "betas=[0.9,0.99]",]
[training_arguments]
pretrained_model_name_or_path = "stablediffusionapi/pony-diffusion-v6-xl"
vae = "stabilityai/sdxl-vae"
max_train_epochs = 10
train_batch_size = 5
seed = 42
max_token_length = 225
xformers = false
sdpa = true
min_snr_gamma = 8.0
lowram = true
no_half_vae = true
gradient_checkpointing = true
gradient_accumulation_steps = 1
max_data_loader_n_workers = 8
persistent_data_loader_workers = true
mixed_precision = "fp16"
full_bf16 = false
cache_latents = true
cache_latents_to_disk = true
cache_text_encoder_outputs = false
min_timestep = 0
max_timestep = 1000
prior_loss_weight = 1.0
[saving_arguments]
save_precision = "fp16"
save_model_as = "safetensors"
save_every_n_epochs = 1
save_last_n_epochs = 10
output_name = "Ruto_Zelda_P1"
output_dir = "/content/drive/MyDrive/lora_training/output/Ruto_Zelda_P1"
log_prefix = "Ruto_Zelda_P1"
logging_dir = "/content/drive/MyDrive/lora_training/log"
FAQ
Comments (10)
Great Model. Almost never breaks, looks good in a variety of different styles.
Thanks!
Ruto realy grow up ¬v¬...
but could be bigger ¬v¬
joke XD
10/10
Would play through the entirety of both Jabu-Jabu and Water Temple per generation if I had to
Water temple music intensifies
LOL Luckily this Lora is easier than the water temple.
Please, can you make this LoRA for Illustrious? Thanks
Yes. I've been meaning to upgrade her for awhile.
I just queued the training. 👍
Sooooo, is illustrious a direct upgrade to pony?
@JustCallMeArbiter
For the record there is nothing wrong with Pony and I'm not going to tell you to stop using Pony but Illustrious is pretty amazing.
Illustrious is more flexible, is more precise, is produces cleaner results than Pony. Its also 10 times easier to train Loras on Illustrious than Pony lol.
Its also much more responsive so it doesnt need as much tags or embedding to prompt it. This makes it hard for some people switching from Pony to Illustrious because they are used to needing a lot of tags and embeddings.
Pony has probably more style loras and way more concepts so Illustrious cant do the crazy concepts that Pony can.
I did make the Illustrious Ruto Lora but I've had a lot of other project get prioritized above it so I havent had a chance to test it.
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.