Update app.py

app.py

@@ -60,7 +60,7 @@ if TORCH_COMPILE:
 pipe.load_lora_weights(
     "lcm-sd/lcm-sdxl-lora",
     weight_name="lcm_sdxl_lora.safetensors",
-    adapter_name="lcm",
+    #adapter_name="lcm",
     use_auth_token=HF_TOKEN,
 )
 
@@ -123,7 +123,8 @@ css = """
 with gr.Blocks(css=css) as demo:
     with gr.Column(elem_id="container"):
         gr.Markdown(
-            """# Ultra-Fast SDXL a Latent Consistency Model LoRA
+            """# Ultra-Fast SDXL with Latent Consistency LoRA
+            In this Space, SDXL is loaded with a latent consistency LoRA, giving it the super power of doing inference in as little as 4 steps. [Learn more on our blog](#) or [technical report](#).
             """,
             elem_id="intro",
         )
@@ -133,6 +134,8 @@ with gr.Blocks(css=css) as demo:
                     placeholder="Insert your prompt here:", value="papercut style of a cute monster", scale=5, container=False
                 )
                 generate_bt = gr.Button("Generate", scale=1)
+        
+        image = gr.Image(type="filepath")
         with gr.Accordion("Advanced options", open=False):
             guidance = gr.Slider(
                 label="Guidance", minimum=0.0, maximum=5, value=0.3, step=0.001
@@ -141,8 +144,23 @@ with gr.Blocks(css=css) as demo:
             seed = gr.Slider(
                 randomize=True, minimum=0, maximum=12013012031030, label="Seed", step=1
             )
-        image = gr.Image(type="filepath")
-
+        with gr.Group():
+            gr.Markdown('''## Using it with `diffusers`
+            ```py
+            from diffusers import DiffusionPipeline, LCMScheduler
+            pipe = DiffusionPipeline.from_pretrained("stabilityai/stable-diffusion-xl-base-1.0").to("cuda") 
+            pipe.scheduler = LCMScheduler.from_config(pipe.scheduler.config)
+            pipe.load_lora_weights("lcm-sd/lcm-sdxl-lora")
+
+            results = pipe(
+                prompt="The spirit of a tamagotchi wandering in the city of Vienna",
+                num_inference_steps=4,
+                guidance_scale=0.5,
+            )
+            results.images[0]
+            ```
+            ''')
+            
         inputs = [prompt, guidance, steps, seed]
         generate_bt.click(fn=predict, inputs=inputs, outputs=image)