import torch
from diffusers import DiffusionPipeline
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("krea/Krea-2-Raw", dtype=torch.bfloat16, device_map="cuda")
pipe.load_lora_weights("quarterturn/krea2-yoimiya")
prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k"
image = pipe(prompt).images[0]A krea2 bf16 LoRA for making Yoimiya Genshin Impact-style images. Use the checkpoint you like best, but I think 9000-10000 is pretty good.
Trained using ai-toolkit CyberHarem/yoimiya_genshin "other" (877 images), captioned using qwen3.6-35-a3b. Checkpoint to step 10000 is provided.
- Downloads last month
- 145
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support
Model tree for quarterturn/krea2-yoimiya
Base model
krea/Krea-2-Raw