update readme

README.md

@@ -30,9 +30,9 @@ Our 1-step Text-to-Image demo is hosted at [DMD2-1step](https://cc2622c0c132346c
 
 We can use the standard diffuser pipeline:
 
-#### 4-step generation 
+#### 4-step UNet generation 
 
-```.bash
+```python
 import torch
 from diffusers import DiffusionPipeline, UNet2DConditionModel, LCMScheduler
 from huggingface_hub import hf_hub_download
@@ -51,9 +51,31 @@ prompt="a photo of a cat"
 image=pipe(prompt=prompt, num_inference_steps=4, guidance_scale=0, timesteps=[999, 749, 499, 249]).images[0]
 ```
 
-#### 1-step generation 
+#### 4-step LoRA generation 
 
-```.bash
+```python
+import torch
+from diffusers import DiffusionPipeline, UNet2DConditionModel, LCMScheduler
+from huggingface_hub import hf_hub_download
+from safetensors.torch import load_file
+base_model_id = "stabilityai/stable-diffusion-xl-base-1.0"
+repo_name = "tianweiy/DMD2"
+ckpt_name = "dmd2_sdxl_4step_lora_fp16.bin"
+# Load model.
+pipe = DiffusionPipeline.from_pretrained(base_model_id, torch_dtype=torch.float16, variant="fp16").to("cuda")
+pipe.load_lora_weights(hf_hub_download(repo, ckpt))
+pipe.fuse_lora()
+
+pipe.scheduler = LCMScheduler.from_config(pipe.scheduler.config)
+prompt="a photo of a cat"
+
+# LCMScheduler's default timesteps are different from the one we used for training 
+image=pipe(prompt=prompt, num_inference_steps=4, guidance_scale=0, timesteps=[999, 749, 499, 249]).images[0]
+```
+
+#### 1-step UNet generation 
+
+```python
 import torch
 from diffusers import DiffusionPipeline, UNet2DConditionModel, LCMScheduler
 from huggingface_hub import hf_hub_download