How to use soloai1/fluxv3 with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("black-forest-labs/FLUX.1-dev", dtype=torch.bfloat16, device_map="cuda") pipe.load_lora_weights("soloai1/fluxv3") prompt = "A mystical top-down view of dnd battle map of an ancient arcane observatory centered around a perfectly circular stone structure with pathways extending in the four cardinal directions. A ring of glowing blue runes encircles the central hub, their eldritch script pulsing with magical energy that matches the numerous small pools of luminescent azure water scattered throughout the surrounding birch forest. The white-trunked trees with their autumnal foliage of oranges and reds create a striking contrast against the verdant ground and mystical blue elements, suggesting a place where the natural and magical worlds converge. Stone paths connect the various ritual sites and glowing fonts, forming a complex that radiates symmetry and purpose—an ideal location for spellcasting, communing with extraplanar entities, or deciphering ancient magical knowledge that has been hidden within this carefully designed arcane focus point since time immemorial." image = pipe(prompt).images[0]
