Enhance CogVideoX Gradio interface with device selection, default parameters, and example prompts

app.py

@@ -20,13 +20,15 @@ def generate_video(transformer_model, prompt, negative_prompt, num_frames, heigh
         transformer_model, 
         torch_dtype=torch.bfloat16
     )
+
+    device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
     
     # Inicializar el pipeline
     pipeline = DiffusionPipeline.from_pretrained(
         "THUDM/CogVideoX-5b", 
         transformer=transformer, 
         torch_dtype=torch.bfloat16
-    ).to("cuda")
+    ).to(device)
     
     # Generar el video
     video_frames = pipeline(
@@ -45,7 +47,7 @@ def generate_video(transformer_model, prompt, negative_prompt, num_frames, heigh
 
 # Crear la interfaz de Gradio
 with gr.Blocks() as demo:
-    gr.Markdown("# Generador de Videos con CogVideoX")
+    gr.Markdown("# CogVideoX Video Generator")
     
     with gr.Row():
         with gr.Column():
@@ -53,59 +55,99 @@ with gr.Blocks() as demo:
             model_dropdown = gr.Dropdown(
                 choices=TRANSFORMER_MODELS,
                 value=TRANSFORMER_MODELS[0],
-                label="Modelo Transformer"
+                label="Transformer Model"
             )
             prompt_input = gr.Textbox(
                 lines=5,
                 label="Prompt",
-                placeholder="Describe el video que quieres generar..."
+                placeholder="Describe the video you want to generate..."
             )
             negative_prompt_input = gr.Textbox(
                 lines=2,
-                label="Prompt Negativo",
+                label="Negative Prompt",
                 value="inconsistent motion, blurry motion, worse quality, degenerate outputs, deformed outputs"
             )
             
-            with gr.Accordion("Parámetros Avanzados", open=False):
+            with gr.Accordion("Advanced Parameters", open=False):
                 num_frames = gr.Slider(
                     minimum=8,
                     maximum=128,
-                    value=8,
+                    value=50,
                     step=1,
-                    label="Número de Frames",
-                    info="Cantidad de frames en el video"
+                    label="Number of Frames",
+                    info="Number of frames in the video"
                 )
                 height = gr.Slider(
                     minimum=32,
                     maximum=1024,
-                    value=32,
+                    value=256,
                     step=64,
-                    label="Altura",
-                    info="Altura del video en píxeles"
+                    label="Height",
+                    info="Video height in pixels"
                 )
                 width = gr.Slider(
                     minimum=32,
                     maximum=1024,
-                    value=32,
+                    value=256,
                     step=64,
-                    label="Anchura",
-                    info="Anchura del video en píxeles"
+                    label="Width",
+                    info="Video width in pixels"
                 )
                 num_inference_steps = gr.Slider(
                     minimum=10,
                     maximum=100,
-                    value=10,
+                    value=50,
                     step=1,
-                    label="Pasos de Inferencia",
-                    info="Mayor número = mejor calidad pero más lento"
+                    label="Inference Steps",
+                    info="Higher number = better quality but slower"
                 )
             
-            generate_btn = gr.Button("Generar Video")
+            generate_btn = gr.Button("Generate Video")
         
         with gr.Column():
             # Salida
-            video_output = gr.Video(label="Video Generado")
+            video_output = gr.Video(label="Generated Video")
     
+    # Agregar ejemplos
+    gr.Examples(
+        examples=[
+            [
+                "sayakpaul/pika-dissolve-v0",
+                "PIKA_DISSOLVE A slender glass vase, brimming with tiny white pebbles, stands centered on a polished ebony dais. Without warning, the glass begins to dissolve from the edges inward. Wisps of translucent dust swirl upward in an elegant spiral, illuminating each pebble as they drop onto the dais. The gently drifting dust eventually settles, leaving only the scattered stones and faint traces of shimmering powder on the stage.",
+                "inconsistent motion, blurry motion, worse quality, degenerate outputs, deformed outputs",
+                8, 32, 32, 10
+            ],
+            [
+                "finetrainers/crush-smol-v0",
+                "DIFF_crush A thick burger is placed on a dining table, and a large metal cylinder descends from above, crushing the burger as if it were under a hydraulic press. The bulb is crushed, leaving a pile of debris around it.",
+                "inconsistent motion, blurry motion, worse quality, degenerate outputs, deformed outputs",
+                8, 32, 32, 10
+            ],
+            [
+                "finetrainers/3dgs-v0",
+                "3D_dissolve In a 3D appearance, a bookshelf filled with books is surrounded by a burst of red sparks, creating a dramatic and explosive effect against a black background.",
+                "inconsistent motion, blurry motion, worse quality, degenerate outputs, deformed outputs",
+                8, 32, 32, 10
+            ],
+            [
+                "finetrainers/cakeify-v0",
+                "PIKA_CAKEIFY On a gleaming glass display stand, a sleek black purse quietly commands attention. Suddenly, a knife appears and slices through the shoe, revealing a fluffy vanilla sponge at its core. Immediately, it turns into a hyper-realistic prop cake, delighting the senses with its playful juxtaposition of the everyday and the extraordinary.",
+                "inconsistent motion, blurry motion, worse quality, degenerate outputs, deformed outputs",
+                8, 32, 32, 10
+            ]
+        ],
+        inputs=[
+            model_dropdown,
+            prompt_input,
+            negative_prompt_input,
+            num_frames,
+            height,
+            width,
+            num_inference_steps
+        ],
+        label="Prompt Examples"
+    )
+
     # Conectar la función
     generate_btn.click(
         fn=generate_video,