v2 [XL]
Higher fidelity images
Better prompt adherence
Better eyes and limbs
This model produces high-quality images in a structured 8-step process.