ComfyUI output: image-to-image workflow with 0.5 (de)noise strength, helloyoung25d_v10f as the diffusion model, and prompt "bakery with a large display window, with many delicious French breads under warm yellow lighting from inside"

Photo