<h3 id="new-sdxl-version-released!-vhm3ngcw4">New SDXL Version released!</h3>I generated the showcase images with TensorArt you can see the details here <a target="_blank" rel="ugc" href="https://tensor.art/models/679231485947122926">https://tensor.art/models/679231485947122926</a>The generation data in the images was manually edited so it might not be perfect but the prompts and resources are correct.<hr />Please checkout my other Game Assets: <a target="_blank" rel="ugc" href="https://civarchive.com/collections/111873">https://civarchive.com/collections/111873</a><hr />Trigger: <code>Isometric_Setting</code>There are infinite combinations of prompts because this can make any setting but here are a few I used. Please check my images for more ideas.<code>cave</code><code>cave, crystal, magic, glowing,</code><code>tropical beach and palm trees, tiny hut, ocean in background,</code><code>ancient ruins, forest in the background,</code><code>paradise, floating island, waterfall, clouds and sky in the background,</code><code>log cabin, snow, pine trees in background,</code>

Stylized Setting (Isometric) SDXL & SD1.5

[sdxl_arguments]cache_text_encoder_outputs = trueno_half_vae = truemin_timestep = 0max_timestep = 1000shuffle_caption = falselowram = true[model_arguments]pretrained_model_name_or_path = "stabilityai/stable-diffusion-xl-base-1.0"vae = "/content/vae/sdxl_vae.safetensors"[dataset_arguments]debug_dataset = falsein_json = "/content/LoRA/meta_lat.json"train_data_dir = "/content/drive/MyDrive/lora_training/datasets/Stylized_Setting_SDXL"dataset_repeats = 2keep_tokens = 1resolution = "1024,1024"color_aug = falsetoken_warmup_min = 1token_warmup_step = 0[training_arguments]output_dir = "/content/drive/MyDrive/kohya-trainer/output/Stylized_Setting_SDXL"output_name = "Stylized_Setting_SDXL"save_precision = "fp16"save_every_n_epochs = 1train_batch_size = 4max_token_length = 225mem_eff_attn = falsesdpa = truexformers = falsemax_train_epochs = 10max_data_loader_n_workers = 8persistent_data_loader_workers = truegradient_checkpointing = truegradient_accumulation_steps = 1mixed_precision = "fp16"[logging_arguments]log_with = "tensorboard"logging_dir = "/content/LoRA/logs"log_prefix = "Stylized_Setting_SDXL"[sample_prompt_arguments]sample_sampler = "euler_a"[saving_arguments]save_model_as = "safetensors"[optimizer_arguments]optimizer_type = "AdaFactor"learning_rate = 0.0001max_grad_norm = 0optimizer_args = [ "scale_parameter=False", "relative_step=False", "warmup_init=False",]lr_scheduler = "constant_with_warmup"lr_warmup_steps = 100[additional_network_arguments]no_metadata = falsenetwork_module = "networks.lora"network_dim = 32network_alpha = 16network_args = []network_train_unet_only = true

Stylized_Setting_SDXL.safetensors

SDXL

v1.0

I'm accepting Lora requests: https://civitai.com/articles/1178Weight: 0.7Trigger: Isometric_Setting &lt;-- Same as the Lora file name, isn't that easy to remember?There are infinite combinations of prompts because this can make any setting but here are a few I used. Please check my images for more ideas.cavecave, crystal, magic, glowing,tropical beach and palm trees, tiny hut, ocean in background,ancient ruins, forest in the background,paradise, floating island, waterfall, clouds and sky in the background,log cabin, snow, pine trees in background,

Files with SHA256: f81b022307c2d7ed8984291a6889096c648f29e419a7a7468a64d3505f27ea87