How can i inference without running the server

[xxivk]

Hello , i ve been working on inference SDXL but i couldnt get same quality as i get in webui

here is my simple code :

`from diffusers import StableDiffusionXLPipeline ,DDIMScheduler ,EulerDiscreteScheduler
import torch

text2img = StableDiffusionXLPipeline.from_single_file(
pretrained_model_link_or_path="https://huggingface.co/ionet-official/bc8-alpha/blob/main/bc8-alpha.safetensors",
use_safetensors=True,
torch_dtype=torch.float16, variant="fp16"

)

Set the scheduler to Euler

#scheduler = DDIMScheduler(beta_schedule="linear", clip_sample=False, set_alpha_to_one=False)
#text2img.scheduler = scheduler

Move to CUDA if available

text2img.to("cuda");

prompt="detailed portrait Neon boxer Brazilian Girl, cyberpunk futuristic neon, reflective crop top and shorts, boxing gloves, decorated with traditional Japanese ornaments by Ismail inceoglu dragan bibin hans thoma greg rutkowski Alexandros Pyromallis Nekro Rene Maritte Illustrated, Perfect face, fine details, realistic shaded, fine-face, pretty face"
negative_prompt ="(worst quality:1.5), (low quality:1.5), (normal quality:1.5), lowres, bad anatomy, bad hands, multiple eyebrow, (cropped), extra limb, missing limbs, deformed hands, long neck, long body, (bad hands), signature, username, artist name, conjoined fingers, deformed fingers, ugly eyes, imperfect eyes, skewed eyes, unnatural face, unnatural body, error, painting by bad-artist"

res = text2img(
prompt=prompt,
negative_prompt=negative_prompt,
width=512,
height=512,
guidance_scale=7,
target_size=(1024,1024),
original_size=(4096,4096),
num_inference_steps=50,
).images[0]

#res = text2img(prompt=prompt)

res
`

how can i do the same way but using SD - WEBUI

am confused the project has much files