Update app.py

app.py

@@ -31,6 +31,7 @@ from ltx_pipelines.utils.constants import (
     DEFAULT_LORA_STRENGTH,
 )
 from ltx_core.loader.single_gpu_model_builder import enable_only_lora
+from PIL import Image
 
 MAX_SEED = np.iinfo(np.int32).max
 # Import from public LTX-2 package
@@ -235,7 +236,7 @@ loras = [
     # --- fused / base behavior ---
     LoraPathStrengthAndSDOps(
         path=distilled_lora_path,
-        strength=0.6,
+        strength=DEFAULT_LORA_STRENGTH,
         sd_ops=LTXV_LORA_COMFY_RENAMING_MAP,
     ),
     LoraPathStrengthAndSDOps(static_lora_path,      DEFAULT_LORA_STRENGTH, LTXV_LORA_COMFY_RENAMING_MAP),
@@ -570,18 +571,20 @@ class CameraDropdown(gr.HTML):
         )
 
         
-def generate_video_example(input_image, prompt, camera_lora, progress=gr.Progress(track_tqdm=True)):
+def generate_video_example(input_image, prompt, camera_lora, resolution, progress=gr.Progress(track_tqdm=True)):
+
+    w, h = apply_resolution(resolution)
 
     output_video, seed = generate_video(
         input_image,
         prompt,
-        5,                      # duration seconds
+        10,                      # duration seconds
         True,                    # enhance_prompt
         42,                      # seed
         True,                    # randomize_seed
-        DEFAULT_1_STAGE_HEIGHT,  # height
-        DEFAULT_1_STAGE_WIDTH,   # width
-        camera_lora,      
+        h,  # height
+        w,   # width
+        camera_lora,     
         progress
     )
 
@@ -714,7 +717,14 @@ def generate_video(
 
 
 def apply_resolution(resolution: str):
-    w, h = resolution.split("x")
+
+    if resolution == "16:9":
+        w, h = 768, 512
+    elif resolution == "1:1":
+        w, h = 512, 512
+    elif resolution == "9:16":
+        w, h = 512, 768
+        
     return int(w), int(h)
 
 def apply_duration(duration: str):
@@ -739,15 +749,6 @@ css = """
     min-width: 0 !important;
     }
 
-    #controls-row #camera_lora_ui {
-    margin-left: auto !important;
-    }
-
-    /* Gradio HTML components often have an inner wrapper div that is width:100% */
-    #camera_lora_ui,
-    #camera_lora_ui > div {
-    width: fit-content !important;
-    }
 
     /* Same idea for your radio HTML blocks (optional but helps) */
     #radioanimated_duration,
@@ -1138,24 +1139,26 @@ with gr.Blocks(title="LTX-2 Video Distilled 🎥🔈") as demo:
             
                 with gr.Row(elem_id="controls-row"):
                         
-                        radioanimated_duration = RadioAnimated(
+                        radioanimated_duration = CameraDropdown(
                             choices=["3s", "5s", "10s", "15s"],
-                            value="3s",
+                            value="5s",
+                            title="Clip Duration",
                             elem_id="radioanimated_duration"
                         )
-                        
+                    
                         duration = gr.Slider(
                             label="Duration (seconds)",
                             minimum=1.0,
                             maximum=15.0,
-                            value=3.0,
+                            value=5.0,
                             step=0.1,
                             visible=False
                         )
 
-                        radioanimated_resolution = RadioAnimated(
-                            choices=["768x512", "512x512", "512x768"],
-                            value=f"{DEFAULT_1_STAGE_WIDTH}x{DEFAULT_1_STAGE_HEIGHT}",
+                        radioanimated_resolution = CameraDropdown(
+                            choices=["16:9", "1:1", "9:16"],
+                            value="16:9",
+                            title="Resolution",
                             elem_id="radioanimated_resolution"
                         )
                         
@@ -1179,6 +1182,28 @@ with gr.Blocks(title="LTX-2 Video Distilled 🎥🔈") as demo:
                     
                 generate_btn = gr.Button("🤩 Generate Video", variant="primary", elem_classes="button-gradient")
 
+    # with gr.Sidebar(width=280):
+    #     gr.Examples(
+    #         examples=[
+    #             [
+    #                 "The video opens on a cake. A knife, held by a hand, is coming into frame and hovering over the cake. The knife then begins cutting into the cake to c4k3 cakeify it. As the knife slices the cake open, the inside of the cake is revealed to be cake with chocolate layers. The knife cuts through and the contents of the cake are revealed. Greek music playing in the background.",
+    #                 "Cakify",
+    #             ],
+
+    #             [
+    #                 "The video showcases an item. The camera zooms out. Then infl4t3 inflates it, the item expanding into giant, inflated balloon against the landscape.",
+    #                 "Inflate",
+    #             ],
+
+    #             [
+    #                 "The video begins with an item. A hydraulic press positioned above slowly descends towards the item. Upon contact, the hydraulic press c5us4 crushes it, deforming and flattening the item, causing the item to collapse inward until the item is no longer recognizable.",
+    #                 "Hydraulic",
+    #             ],
+    #         ],
+    #         inputs=[prompt_ui, camera_lora_ui],
+    #         label="Example",
+    #         cache_examples=False,
+    #     )
     camera_lora_ui.change(
         fn=lambda x: x,
         inputs=camera_lora_ui,
@@ -1229,26 +1254,45 @@ with gr.Blocks(title="LTX-2 Video Distilled 🎥🔈") as demo:
                 "supergirl.png",
                 "A fuzzy puppet superhero character resembling a female puppet with blonde hair and a blue superhero suit stands inside an icy cave made of frozen walls and icicles, she looks panicked and frantic, rapidly turning her head left and right and scanning the cave while waving her arms and shouting angrily and desperately, mouthing the words “where the hell is my dog,” her movements exaggerated and puppet-like with high energy and urgency, suddenly a second puppet dog bursts into frame from the side, jumping up excitedly and tackling her affectionately while licking her face repeatedly, she freezes in surprise and then breaks into relief and laughter as the dog continues licking her, the scene feels chaotic, comedic, and emotional with expressive puppet reactions, cinematic lighting, smooth camera motion, shallow depth of field, and high-quality puppet-style animation",
                 "No LoRA",
+                "16:9",
             ],
             [
                 "highland.png",
                 "Realistic POV selfie-style video in a snowy, foggy field. Two shaggy Highland cows with long curved horns stand ahead. The camera is handheld and slightly shaky. The woman filming talks nervously and excitedly in a vlog tone: \"Oh my god guys… look how big those horns are… I’m kinda scared.\" The cow on the left walks toward the camera in a cute, bouncy, hopping way, curious and gentle. Snow crunches under its hooves, breath visible in the cold air. The horns look massive from the POV. As the cow gets very close, its wet nose with slight dripping fills part of the frame. She laughs nervously but reaches out and pets the cow. The cow makes deep, soft, interesting mooing and snorting sounds, calm and friendly. Ultra-realistic, natural lighting, immersive audio, documentary-style realism.",
                 "No LoRA",
+                "16:9",
             ],
             [
                 "wednesday.png",
                 "A cinematic close-up of Wednesday Addams frozen mid-dance on a dark, blue-lit ballroom floor as students move indistinctly behind her, their footsteps and muffled music reduced to a distant, underwater thrum; the audio foregrounds her steady breathing and the faint rustle of fabric as she slowly raises one arm, never breaking eye contact with the camera, then after a deliberately long silence she speaks in a flat, dry, perfectly controlled voice, “I don’t dance… I vibe code,” each word crisp and unemotional, followed by an abrupt cutoff of her voice as the background sound swells slightly, reinforcing the deadpan humor, with precise lip sync, minimal facial movement, stark gothic lighting, and cinematic realism.",
                  "Dolly Out",
+                "16:9",
             ],
             [
                 "astronaut.png",
                 "An astronaut hatches from a fragile egg on the surface of the Moon, the shell cracking and peeling apart in gentle low-gravity motion. Fine lunar dust lifts and drifts outward with each movement, floating in slow arcs before settling back onto the ground. The astronaut pushes free in a deliberate, weightless motion, small fragments of the egg tumbling and spinning through the air. In the background, the deep darkness of space subtly shifts as stars glide with the camera's movement, emphasizing vast depth and scale. The camera performs a smooth, cinematic slow push-in, with natural parallax between the foreground dust, the astronaut, and the distant starfield. Ultra-realistic detail, physically accurate low-gravity motion, cinematic lighting, and a breath-taking, movie-like shot.",
                 "Static",
-            ]
+                "1:1",
+            ],
+            # [
+            #     "astronaut.png",
+            #     "The video opens on object. A knife, held by a hand, is coming into frame and hovering over the object. The knife then begins cutting into the object to c4k3 cakeify them. As the knife slices the object open, the insides of the object are revealed to be cakes with chocolate layers. The knife cuts through and the contents of object are revealed.",
+            #     "Cakify",
+            # ],
+            # [
+            #     "astronaut.png",
+            #     "The video showcases an item. The camera zooms out. Then infl4t3 inflates it, the item expanding into giant, inflated balloon against the landscape.",
+            #     "Inflate",
+            # ],
+            # [
+            #     "astronaut.png",
+            #     "The video begins with an item. A hydraulic press positioned above slowly descends towards the item. Upon contact, the hydraulic press c5us4 crushes it, deforming and flattening the item, causing the item to collapse inward until the item is no longer recognizable.",
+            #     "Hydraulic",
+            # ],
             
         ],
         fn=generate_video_example,
-        inputs=[input_image, prompt_ui, camera_lora_ui],
+        inputs=[input_image, prompt_ui, camera_lora_ui, radioanimated_resolution],
         outputs = [output_video],
         label="Example",
         cache_examples=True,