ProGamerGov/360-Diffusion-LoRA-sd-v1-5 · Cartoonic Textures with Low Detail

Big thanks to ProGamerGov for this contribution. I'm trying it out with the following parameters:

prompt = "Photorealistic image of a sunlit forest with a small calm stream flowing through it, tall trees with lush green foliage, lightly cloudy blue sky above the trees, and dappled sunlight on the forest floor, cinematic lighting, detailed textures, hyperrealistic."
num_inference_steps = 50
seed = 42
height = 512
width = height * 2

The image generated is always cartoonic with very low detail to the textures. I've also tried with 100 inference steps, still the same result. In comparison, the SD1.5 is generating a stunning photograph with the same parameters.

Stable Diffusion v1.5 Default:

Stable Diffusion v1.5 with 360-Diffusion Lora Weights:

Is this expected, or am I doing something wrong? Any help is appreciated.