fix rho in the examples

app.py

@@ -19,7 +19,6 @@ import argparse
 precision = get_precision()
 transformer = NunchakuFluxTransformer2dModel.from_pretrained(f"nunchaku-tech/nunchaku-flux.1-schnell/svdq-{precision}_r32-flux.1-schnell.safetensors")
 pipe = FluxPipeline.from_pretrained("black-forest-labs/FLUX.1-schnell", transformer=transformer, torch_dtype=torch.bfloat16).to("cuda")
-# pipe.vae = AutoencoderTiny.from_pretrained("madebyollin/taef1").to("cuda")
 pipe.vae = AutoencoderTiny.from_pretrained("madebyollin/taef1", torch_dtype=torch.bfloat16).to("cuda")
 
 m_clip = CLIPModel.from_pretrained("multimodalart/clip-vit-base-patch32").to("cuda")
@@ -157,7 +156,7 @@ def get_score_functions(unary_term, binary_term, prompt):
     return unary_score_fn, binary_score_fn
 
 
-@spaces.GPU(duration=300)
+@spaces.GPU(duration=200)
 def generate_images(prompt, starting_candidates, output_group_size, pruning_ratio, 
                    lambda_score, seed, unary_term, binary_term, progress=gr.Progress(track_tqdm=True)):
     """Generate images using group inference with progressive pruning."""
@@ -283,10 +282,10 @@ with gr.Blocks(css=custom_css, js=js_func, theme=gr.themes.Soft(), elem_id="main
     
     gr.Examples(
         examples=[
-            ["A photo of a dog", 64, 4, 0.5, 1.0, 42, "clip_text_img", "diversity_dino"],
-            ["A mountain landscape", 64, 4, 0.5, 1.0, 123, "clip_text_img", "diversity_dino"],
-            ["A cat sleeping", 64, 4, 0.5, 1.0, 456, "clip_text_img", "diversity_dino"],
-            ["A sunset at the beach", 64, 4, 0.5, 1.0, 789, "clip_text_img", "diversity_dino"],
+            ["A photo of a dog", 32, 4, 0.9, 1.0, 42, "clip_text_img", "diversity_dino"],
+            ["A mountain landscape", 32, 4, 0.9, 1.0, 123, "clip_text_img", "diversity_dino"],
+            ["A cat sleeping", 32, 4, 0.9, 1.0, 456, "clip_text_img", "diversity_dino"],
+            ["A sunset at the beach", 32, 4, 0.9, 1.0, 789, "clip_text_img", "diversity_dino"],
         ],
         inputs=[prompt, starting_candidates, output_group_size, pruning_ratio, lambda_score, seed, unary_term, binary_term],
         outputs=[output_gallery_group],