Update app.py

app.py

@@ -1,85 +1,96 @@
 import gradio as gr
-import numpy as np
 import random
 import torch
 from diffusers import DiffusionPipeline
 
-# Check for GPU availability
+# Device setup
 device = "cuda" if torch.cuda.is_available() else "cpu"
-torch_dtype = torch.float16 if torch.cuda.is_available() else torch.float32
+model_repo_id_turbo = "stabilityai/sdxl-turbo"  # Stability AI Model
+pipe_turbo = DiffusionPipeline.from_pretrained(model_repo_id_turbo, torch_dtype=torch.float16 if torch.cuda.is_available() else torch.float32).to(device)
 
-# Load your DiffusionPipeline model
-model_repo_id = "stabilityai/sdxl-turbo"
-pipe = DiffusionPipeline.from_pretrained(model_repo_id, torch_dtype=torch_dtype)
-pipe = pipe.to(device)
+# Placeholder for ZB-Tech model
+def load_zb_model():
+    return gr.Interface.load("models/ZB-Tech/Text-to-Image")
 
-MAX_SEED = np.iinfo(np.int32).max
-MAX_IMAGE_SIZE = 1024
-
-# Define the custom model inference function
+# Inference function
 def custom_infer(
-    prompt, negative_prompt, seed, randomize_seed, width, height, guidance_scale, num_inference_steps
+    model_choice, prompt, negative_prompt, seed, randomize_seed, width, height, guidance_scale, num_inference_steps
 ):
-    if randomize_seed:
-        seed = random.randint(0, MAX_SEED)
-
-    generator = torch.Generator().manual_seed(seed)
-
-    image = pipe(
-        prompt=prompt,
-        negative_prompt=negative_prompt,
-        guidance_scale=guidance_scale,
-        num_inference_steps=num_inference_steps,
-        width=width,
-        height=height,
-        generator=generator,
-    ).images[0]
-
-    return image, seed
-
-
-# Gradio interface for custom model
-def custom_model_ui():
-    with gr.Blocks() as custom_demo:
-        gr.Markdown("## Custom Model: Stability AI SDXL")
-        with gr.Row():
-            prompt = gr.Text(label="Prompt")
-            run_button = gr.Button("Generate")
+    # Load the selected model
+    if model_choice == "Faster image generation (suitable for CPUs)":
+        model = load_zb_model()
+        return model(prompt)
+    else:
+        default_negative_prompt = "no watermark, hezzy, blurry"
+        combined_negative_prompt = f"{default_negative_prompt}, {negative_prompt}" if negative_prompt else default_negative_prompt
+        
+        if randomize_seed:
+            seed = random.randint(0, np.iinfo(np.int32).max)
+        
+        generator = torch.Generator().manual_seed(seed)
+        image = pipe_turbo(
+            prompt=prompt,
+            negative_prompt=combined_negative_prompt,
+            guidance_scale=guidance_scale,
+            num_inference_steps=num_inference_steps,
+            width=width,
+            height=height,
+            generator=generator,
+        ).images[0]
+        return image, seed
+
+# CSS for centering UI
+css = """
+#col-container {
+    display: flex;
+    flex-direction: column;
+    align-items: center;
+    justify-content: center;
+    text-align: center;
+    margin: 0 auto;
+}
+"""
+
+# Gradio app
+with gr.Blocks(css=css) as demo:
+    with gr.Column(elem_id="col-container"):
+        # App name and description
+        gr.Markdown(
+            """
+            # AI-Powered Text-to-Image Generator  
+            *Generate stunning images from text prompts using advanced AI models.*
+            """
+        )
 
-        result = gr.Image(label="Generated Image")
-        negative_prompt = gr.Text(label="Negative Prompt", placeholder="Optional")
-        seed = gr.Slider(0, MAX_SEED, label="Seed", step=1, value=0)
-        randomize_seed = gr.Checkbox(label="Randomize Seed", value=True)
-        width = gr.Slider(256, MAX_IMAGE_SIZE, step=32, value=1024, label="Width")
-        height = gr.Slider(256, MAX_IMAGE_SIZE, step=32, value=1024, label="Height")
-        guidance_scale = gr.Slider(0, 10, step=0.1, value=7.5, label="Guidance Scale")
-        num_inference_steps = gr.Slider(1, 50, step=1, value=30, label="Inference Steps")
+        # Dropdown for model selection
+        model_choice = gr.Dropdown(
+            label="Select Model",
+            choices=[
+                "Faster image generation (suitable for CPUs)",
+                "More customizable option (slower, suitable for GPUs)"
+            ],
+            value="Faster image generation (suitable for CPUs)",
+        )
 
-        run_button.click(
+        # Input section
+        prompt = gr.Textbox(label="Prompt", placeholder="Enter your prompt here...")
+        with gr.Accordion("Advanced Settings", open=False):
+            negative_prompt = gr.Textbox(label="Negative Prompt", placeholder="Enter a negative prompt here...")
+            seed = gr.Slider(label="Seed", minimum=0, maximum=2147483647, step=1, value=0)
+            randomize_seed = gr.Checkbox(label="Randomize Seed", value=True)
+            width = gr.Slider(label="Width", minimum=256, maximum=1024, step=32, value=512)
+            height = gr.Slider(label="Height", minimum=256, maximum=1024, step=32, value=512)
+            guidance_scale = gr.Slider(label="Guidance Scale", minimum=0.0, maximum=10.0, step=0.1, value=7.5)
+            num_inference_steps = gr.Slider(label="Inference Steps", minimum=1, maximum=50, step=1, value=25)
+
+        # Output section
+        result = gr.Image(label="Generated Image", type="pil")
+        gr.Button("Generate").click(
             custom_infer,
-            inputs=[prompt, negative_prompt, seed, randomize_seed, width, height, guidance_scale, num_inference_steps],
-            outputs=[result, seed],
+            inputs=[model_choice, prompt, negative_prompt, seed, randomize_seed, width, height, guidance_scale, num_inference_steps],
+            outputs=result
         )
 
-    return custom_demo
-
-
-# Preloaded Gradio model
-def preloaded_model_ui():
-    with gr.Blocks() as preloaded_demo:
-        gr.Markdown("## Preloaded Model: ZB-Tech Text-to-Image")
-        preloaded_demo = gr.load("models/ZB-Tech/Text-to-Image")
-
-    return preloaded_demo
-
-
-# Combine both interfaces in tabs
-with gr.Blocks() as demo:
-    with gr.Tab("Custom Model"):
-        custom_ui = custom_model_ui()
-
-    with gr.Tab("Preloaded Model"):
-        preloaded_ui = preloaded_model_ui()
-
+# Launch app
 if __name__ == "__main__":
     demo.launch()