Update app.py

app.py

@@ -5,20 +5,16 @@ from diffusers import DiffusionPipeline
 from optimum.intel.openvino import OVStableDiffusionPipeline
 import torch
 
-device = "cuda" if torch.cuda.is_available() else "cpu"
+model_id = "helenai/Linaqruf-anything-v3.0-ov"
+
+pipe = OVStableDiffusionPipeline.from_pretrained(model_id, compile=False)
+pipe.reshape( batch_size=batch_size, height=height, width=width, num_images_per_prompt=num_images_per_prompt)
+pipe.compile()
+
 
-if torch.cuda.is_available():
-    torch.cuda.max_memory_allocated(device=device)
-    pipe = DiffusionPipeline.from_pretrained("stabilityai/sdxl-turbo", torch_dtype=torch.float16, variant="fp16", use_safetensors=True)
-    pipe.enable_xformers_memory_efficient_attention()
-    pipe = pipe.to(device)
-else: 
-    #pipe = DiffusionPipeline.from_pretrained("stabilityai/sdxl-turbo", use_safetensors=True)
-    pipe = OVStableDiffusionPipeline.from_pretrained("helenai/Linaqruf-anything-v3.0-ov")
-    pipe = pipe.to(device)
 
 MAX_SEED = np.iinfo(np.int32).max
-MAX_IMAGE_SIZE = 1024
+MAX_IMAGE_SIZE = 256
 
 def infer(prompt, negative_prompt, seed, randomize_seed, width, height, guidance_scale, num_inference_steps):
 
@@ -52,10 +48,8 @@ css="""
 }
 """
 
-if torch.cuda.is_available():
-    power_device = "GPU"
-else:
-    power_device = "CPU"
+
+power_device = "CPU"
 
 with gr.Blocks(css=css) as demo: