Trained upon cinematic images taken from World of Warcraft trailers up till the "Shadowlands" expansion. This is also why it seems to produce better results in landscape than in square and portrait. I got the best results with "Euler a" 32 steps 7.5 cfg + Refiner and/or Hi-Res fix (to avoid pixelated results). Append your prompts with "World of Warcraft cinematic style" to trigger the style

Trigger words

You should use World of Warcraft cinematic style to trigger the image generation.

Download model

Weights for this model are available in Safetensors format.

Download them in the Files & versions tab.

Use it with the 🧨 diffusers library

from diffusers import AutoPipelineForText2Image
import torch

pipeline = AutoPipelineForText2Image.from_pretrained('stabilityai/stable-diffusion-xl-base-1.0', torch_dtype=torch.float16).to('cuda')
pipeline.load_lora_weights('Norod78/world-of-warcraft-cinematic-style-sdxl', weight_name='SDXL-World-of-Warcraft-cinematic-style-Lora-r32-000005.safetensors')
image = pipeline('benderbot World of Warcraft cinematic style   ').images[0]

For more details, including weighting, merging and fusing LoRAs, check the documentation on loading LoRAs in diffusers