Name: CAT - Citron Styles
Rating: 5 (6855 reviews)
Author: CitronLegacy

Any checkpoint can now be in the generator. Anything already loaded works for everyone; members load the new ones free for now, with a faster lane and Boost to jump the queue.

Any Model, Any Time: Load It Into the Generator Yourself

We rebuilt the trainer. Watch your run as it trains, generate with an epoch before it's done, and compare epochs side by side. It's in beta, and you switch it on yourself in your preferences.

Training Studio: Watch It Train, Generate As It Goes

A training and content contest with over 2 Million Buzz in prizes and 50% off H3 generation for the duration of the contest.

The Civitai H3 Contest: 2 Million Buzz Up For Grabs

Follow me on X/Twitter for exclusive content: <a target="_blank" rel="ugc" href="https://x.com/Citron_Legacy">https://x.com/Citron_Legacy</a> 😁Style <a target="_blank" rel="ugc" href="https://civitai.com/models/362745?modelVersionId=2474947">Couture Craze returns on ZImageTurbo</a>!

Trained using the dataset from the Pony version enhanced with images from Flux. Shout out to <a target="_blank" rel="ugc" href="https://civitai.com/user/iJWiTGS8">iJWiTGS8 </a>for helping me learn how to create this style!

Real Anime IXL

C0utur3Cr4z3,&lt;lora:CoutureCraze_IXL:1.0&gt;,This model specializes in clothing styles, trained on over 100 detailed and stylish anime outfits to enhance clothing with intricate details."Couture" refers to the design and creation of high-end, custom-made fashion. It is often associated with haute couture, which means "high sewing" in French, representing exclusive, handcrafted garments tailored to individual clients.

Couture Craze IL

I added some suggested prompts in the Trigger but you don't actually need them.--You can find some of my training data on several of my Lora posts like this one <a target="_blank" rel="ugc" href="https://civitai.com/images/8452326">https://civitai.com/images/8452326</a>--Shout out to <a target="_blank" rel="ugc" href="https://civitai.com/user/iJWiTGS8">iJWiTGS8 </a>for helping me learn how to create this style!--Trained on: 210 imagesSteps: 840Total Time: 2:36:04--[network_arguments]unet_lr = 0.75text_encoder_lr = 0.75network_dim = 32network_alpha = 32network_module = "networks.lora"network_train_unet_only = false[optimizer_arguments]learning_rate = 0.75lr_scheduler = "cosine_with_restarts"lr_scheduler_num_cycles = 3lr_warmup_steps = 42optimizer_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 = 10train_batch_size = 5seed = 42max_token_length = 225xformers = falsesdpa = truemin_snr_gamma = 8.0lowram = trueno_half_vae = truegradient_checkpointing = truegradient_accumulation_steps = 1max_data_loader_n_workers = 8persistent_data_loader_workers = truemixed_precision = "fp16"full_bf16 = falsecache_latents = truecache_latents_to_disk = truecache_text_encoder_outputs = falsemin_timestep = 0max_timestep = 1000prior_loss_weight = 1.0[saving_arguments]save_precision = "fp16"save_model_as = "safetensors"save_every_n_epochs = 1save_last_n_epochs = 10output_name = "RealisticAnime"output_dir = "/content/drive/MyDrive/lora_training/output/RealisticAnime"log_prefix = "RealisticAnime"logging_dir = "/content/drive/MyDrive/lora_training/log"

Realistic Anime

Digital ghosts in static cry,Dreams of dawn refuse to die.In the city submerged in blight,The hero arrives during the White Night!WNS,&lt;lora:WhiteNightStyle_IXL:1.0&gt;,

White Night

masterpiece, best quality, anime coloring, anime screencap,&lt;lora:CoutureCraze_ZIT:1.0&gt;,

CoutureCraze ZIT

MoonlightCAT ZIT

<a target="_blank" rel="ugc" href="https://civitai.com/models/362745?modelVersionId=1824704">Kawaii Mirai</a> (Cute Future)<s>In the grim darkness of the far future...</s> In the cheerful light of the far future, there is only Pastel! CatHammer40K! The Empurror Puurtects! 🐈😜.k4w4iim1r41, cute outfit, pastel,&lt;lora:KawaiiMirai_IXL:1.0&gt;,

Kawaii Mirai IXL

New Version: <a target="_blank" rel="ugc" href="https://civitai.com/models/362745?modelVersionId=2055126">Equipment Install</a>This Lora will add equipment to your character even if you don't prompt it! If you do prompt it then you can get lots of nice details.

Equipment Install IL

Bzzm3t0p, Bzzm3b0t, ,&lt;lora:BuzzMe_IXL:1.0&gt;,

Buzz Me

L1P5, lips focus, ,&lt;lora:L1p5Style_IXL:1.0&gt;,

Lips

It works best for <code>upper body</code> images. (You may need Face fix/Adetailer if you generate cowboy shot images)--zzHyp3r,upper body, backlighting,&lt;lora:HyperHighlight_IXL:1.0&gt;,

Hyper Highlight IL

ProAnime

CuteToon

BeautifulCAT

NebulaNoir

EnergyCAT

Alola

Trained on images generated with my <a target="_blank" rel="ugc" href="https://civitai.com/models/291270">CATalyst Checkpoint</a>--Today my AI Birthday! 🎉I started creating AI models on July 2nd 2023. Lets party! ✨<a target="_blank" rel="ugc" href="https://civitai.com/models/362745?modelVersionId=575354">Post images in my CATaylst Style to join the party.</a>Prize #1 - 2000 Buzz ⚡- 3 WinnersOn July 13th, the top 3 images (rated PG, PG-13, or R) will each win 2000 Buzz.Prize #2 - 100 Buzz ⚡- 10 WinnersThe first 10 users to post images (rated PG, PG-13, or R) will each win 100 Buzz.Extras - 🎲? Buzz ⚡- Unlimited WinnersI’ll tip small amounts of Buzz for every good post (any maturity rating). Users can earn Buzz for multiple good posts.Detailed Rules:1. Prize #1 winning images must be made on the CivitAI Generator and reproducible.2. The top 3 images are those with the most reactions as the first image in a post.3. Users can only win once for Prizes #1 and #2 but can win both Prizes #1 and #2. You could get Prize #2 and then later win Prize #1 but you can’t win Prize #1 twice.4. Users that posted here before the competition will not count towards the first 10 for Prize #2 but they are in the running for Prize #1.5. Spamming posts won’t earn more Buzz, but consistently posting good images will. I’m generous but don't expect to earn 1000 Buzz for 100 posts in one day.6. I, Citron Legacy, cannot win any of these prizes so don’t worry about trying to out preform my images lol7. Feel free to farm for votes I don't consider this cheating8. I reserve the right to change the rules if I feel like someone is cheating the system or my rules were not fair to the participants.

CATalyst

Trained on images generated with <a rel="ugc" href="https://civitai.com/models/131986?modelVersionId=145126">CAT - Citron Anime Treasure</a> with the goal to copy this style to PonyXL--Trained on: 120 imagesSteps: 434/620 Epoch: 7/10Time for Epoch 7: 1:00:47Colab Trainer crashed at:epoch 8/10steps: 77% 480/620 [1:07:15&lt;19:37]--[network_arguments]unet_lr = 0.75text_encoder_lr = 0.75network_dim = 32network_alpha = 32network_module = "networks.lora"network_train_unet_only = false[optimizer_arguments]learning_rate = 0.75lr_scheduler = "cosine_with_restarts"lr_scheduler_num_cycles = 3lr_warmup_steps = 30optimizer_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 = 10train_batch_size = 4seed = 42max_token_length = 225xformers = falsesdpa = truemin_snr_gamma = 8.0lowram = trueno_half_vae = truegradient_checkpointing = truegradient_accumulation_steps = 1max_data_loader_n_workers = 8persistent_data_loader_workers = truemixed_precision = "fp16"full_bf16 = falsecache_latents = truecache_latents_to_disk = truecache_text_encoder_outputs = falsemin_timestep = 0max_timestep = 1000prior_loss_weight = 1.0[saving_arguments]save_precision = "fp16"save_model_as = "safetensors"save_every_n_epochs = 1save_last_n_epochs = 10output_name = "CitronAnimeTreasure"output_dir = "/content/drive/MyDrive/lora_training/output/CitronAnimeTreasure"log_prefix = "CitronAnimeTreasure"logging_dir = "/content/drive/MyDrive/lora_training/log"

Anime Toon

Candy Cute

tinted eyewear,dress, bare shoulders, jewelry, closed mouth, standing, earrings, sleeveless, black dress, bare legs, sleeveless dress, side slit, hoop earrings, pink high heels, navel, jewelry, closed mouth, standing, swimsuit, full body, bikini, earrings, bracelet, sandals, multicolored clothes, multicolored bikini, text, text in background,skirt, shirt, bare shoulders, jewelry, collarbone, short sleeves, cowboy shot, pleated skirt, earrings, striped, off shoulder, makeup, white skirt, off-shoulder shirt, shirt tucked in, multicolored shirt,jewelry, white shirt, upper body, short sleeves, earrings, glasses, striped, hoop earrings, round eyewear, yellow-framed eyewear, shirt, holding, jewelry, sitting, full body, short sleeves, earrings, shorts, striped, collared shirt, bracelet, short shorts, bare legs, black shorts, sandals, microphone, shirt tucked in, holding microphone, high-waist shorts,

80s_Pop

3D Animation

Trained on CivitAI TrainerTrained on: 83 imagesNum of Repeats: 4Number of Epochs: 10Training Batch Size: 4Total Steps: 830Network Dim: 32Network Alpha: 32Keep Tokens: 0Shuffle caption: false

Flat Color

Two of my Loras merged:Vivid = <a target="_blank" rel="ugc" href="https://civitai.com/models/362745?modelVersionId=420176">Realistic Anime</a>:0.5 + <a target="_blank" rel="ugc" href="https://civitai.com/models/185558?modelVersionId=346951">Zankuro</a>:1.0

Vivid

This is heavily, but not entirely, influenced by The Other Half.Please support the artist by checking out their page🔞: <a target="_blank" rel="ugc" href="https://twitter.com/ToHDraws">https://twitter.com/ToHDraws</a>--Trained on the CivitAI trainer.Trained on: 107 imagesTraining Model: ponyDiffusionV6XL_v6StartWithThisOneflip_aug: FalseNum of Repeats: 5Number of Epochs: 10Training Batch Size: 4Total Steps: 1540Resolution: 1024Network Dim: Dim 32 and Alpha 32Optimizer: prodigyopt.prodigy.Prodigy(decouple=True,weight_decay=0.5,betas=(0.9, 0.99),use_bias_correction=False)Training Time: 01:05:43 (Submit time to Ready time)Buzz Cost: 652

Bold Toon

Trained on dataset from my previous Lora <a target="_blank" rel="ugc" href="https://civitai.com/models/202764">Anime Sketch SDXL</a>

Sketch

Trained on the dataset from my other Lora <a target="_blank" rel="ugc" href="https://civitai.com/models/344155/">https://civitai.com/models/344155/</a>

Semi 3D

Merge of two Loras<a rel="ugc" href="https://civitai.com/models/185558?modelVersionId=399313">Zankuro</a>:0.4 + <a rel="ugc" href="https://civitai.com/models/264290?modelVersionId=298238">Styles for Pony Diffusion V6 XL</a>:0.6

Bold Lines

The base of this Lora was trained with the CivitAI trainer on the dataset from my Lora <a target="_blank" rel="ugc" href="https://civitai.com/models/344155">Beautiful CAT</a> (207 images) then I merged it with my <a target="_blank" rel="ugc" href="https://civitai.com/models/185558?modelVersionId=399313">Zankuro Pony</a> LoraI comment the json for this ComfyUI workflow in an image my my workflow: <a target="_blank" rel="ugc" href="https://civitai.com/images/8374468">https://civitai.com/images/8374468</a>