Update README.md

README.md

@@ -74,3 +74,47 @@ Efficient fine-tuning: The Mobius base model serves as a superior foundation for
 
 - a CFG of either 3.5 to 7
 - Requires a CLIP skip of -3
+- Sampler: DPM++ 3M SDE
+- Scheduler: Karras
+- Resolution: 1024x1024
+
+please also consider using these keep words to improve your prompts: best quality, HD, ~*~aesthetic~*~.
+
+
+Use it with 🧨 diffusers
+```
+import torch
+from diffusers import (
+    StableDiffusionXLPipeline, 
+    KDPM2AncestralDiscreteScheduler,
+    AutoencoderKL
+)
+
+# Load VAE component
+vae = AutoencoderKL.from_pretrained(
+    "madebyollin/sdxl-vae-fp16-fix", 
+    torch_dtype=torch.float16
+)
+
+# Configure the pipeline
+pipe = StableDiffusionXLPipeline.from_pretrained(
+    "Corcelio/mobius", 
+    vae=vae,
+    torch_dtype=torch.float16
+)
+pipe.scheduler = KDPM2AncestralDiscreteScheduler.from_config(pipe.scheduler.config)
+pipe.to('cuda')
+
+# Define prompts and generate image
+prompt = "mystery"
+negative_prompt = ""
+
+image = pipe(
+    prompt, 
+    negative_prompt=negative_prompt, 
+    width=1024,
+    height=1024,
+    guidance_scale=7,
+    num_inference_steps=50
+).images[0]
+```