Initial Setup

app.py

@@ -1,7 +1,209 @@
 import gradio as gr
+import torch
+import os
+import spaces
+import uuid
 
-def greet(name):
-    return "Hello " + name + "!!"
+from diffusers import AnimateDiffPipeline, MotionAdapter, EulerDiscreteScheduler
+from diffusers.utils import export_to_video
+from huggingface_hub import hf_hub_download
+from safetensors.torch import load_file
+from PIL import Image
+from transformers import CLIPFeatureExtractor
 
-demo = gr.Interface(fn=greet, inputs="text", outputs="text")
-demo.launch()
+# Custom message for linking to other demos
+MORE_INFO = """ ## Related Demos
+        ### Check out other interesting demos here
+        """
+
+# Model configuration
+BASES = {
+    "Cartoon": "frankjoshua/toonyou_beta6",
+    "Realistic": "emilianJR/epiCRealism", 
+    "3d": "Lykon/DreamShaper",
+    "Anime": "Yntec/mistoonAnime2"
+}
+
+# Global state tracking
+step_loaded = None
+base_loaded = "Realistic"
+motion_loaded = None
+
+# GPU check and initialization
+if not torch.cuda.is_available():
+    raise NotImplementedError("No GPU detected! This space requires a GPU to run.")
+
+device = "cuda"
+dtype = torch.float16
+
+# Initialize the pipeline
+pipe = AnimateDiffPipeline.from_pretrained(
+    BASES[base_loaded], 
+    torch_dtype=dtype
+).to(device)
+
+# Configure scheduler
+pipe.scheduler = EulerDiscreteScheduler.from_config(
+    pipe.scheduler.config, 
+    timestep_spacing="trailing", 
+    beta_schedule="linear"
+)
+
+# Initialize safety checker
+feature_extractor = CLIPFeatureExtractor.from_pretrained("openai/clip-vit-base-patch32")
+
+# Main generation function
+@spaces.GPU(duration=30, queue=False)
+def generate_image(prompt, base="Realistic", motion="", step=8, progress=gr.Progress()):
+    global step_loaded
+    global base_loaded
+    global motion_loaded
+    print(f"Generating with prompt: {prompt}, base: {base}, steps: {step}")
+
+    # Load step configuration if needed
+    if step_loaded != step:
+        repo = "ByteDance/AnimateDiff-Lightning"
+        ckpt = f"animatediff_lightning_{step}step_diffusers.safetensors"
+        pipe.unet.load_state_dict(
+            load_file(hf_hub_download(repo, ckpt), device=device), 
+            strict=False
+        )
+        step_loaded = step
+
+    # Load base model if needed
+    if base_loaded != base:
+        pipe.unet.load_state_dict(
+            torch.load(
+                hf_hub_download(BASES[base], "unet/diffusion_pytorch_model.bin"), 
+                map_location=device
+            ), 
+            strict=False
+        )
+        base_loaded = base
+
+    # Handle motion adaptation
+    if motion_loaded != motion:
+        pipe.unload_lora_weights()
+        if motion != "":
+            pipe.load_lora_weights(motion, adapter_name="motion")
+            pipe.set_adapters(["motion"], [0.7])
+        motion_loaded = motion
+
+    # Progress tracking
+    progress((0, step))
+    def progress_callback(i, t, z):
+        progress((i+1, step))
+
+    # Generate the video
+    output = pipe(
+        prompt=prompt,
+        guidance_scale=1.2,
+        num_inference_steps=step,
+        callback=progress_callback,
+        callback_steps=1
+    )
+
+    # Save and return the video
+    name = str(uuid.uuid4()).replace("-", "")
+    path = f"/tmp/{name}.mp4"
+    export_to_video(output.frames[0], path, fps=10)
+    return path
+
+# Gradio Interface
+with gr.Blocks(css="style.css") as demo:
+    # Header
+    gr.HTML(
+        """
+        <h1><center>Instant⚡Video Generator</center></h1>
+        <p><center><span class='warning'>Adjust steps from 4 to 8 for better results if needed.</span></center></p>
+        <p><center><strong>First generation takes longer, subsequent generations are faster.</strong></p>
+        <p><center>Follow the example prompts format for best results</p>
+        """
+    )
+
+    # Input interface
+    with gr.Group():
+        with gr.Row():
+            prompt = gr.Textbox(label='Prompt')
+        
+        with gr.Row():
+            select_base = gr.Dropdown(
+                label='Base model',
+                choices=list(BASES.keys()),
+                value=base_loaded,
+                interactive=True
+            )
+            
+            select_motion = gr.Dropdown(
+                label='Motion',
+                choices=[
+                    ("Default", ""),
+                    ("Zoom in", "guoyww/animatediff-motion-lora-zoom-in"),
+                    ("Zoom out", "guoyww/animatediff-motion-lora-zoom-out"),
+                    ("Tilt up", "guoyww/animatediff-motion-lora-tilt-up"),
+                    ("Tilt down", "guoyww/animatediff-motion-lora-tilt-down"),
+                    ("Pan left", "guoyww/animatediff-motion-lora-pan-left"),
+                    ("Pan right", "guoyww/animatediff-motion-lora-pan-right"),
+                    ("Roll left", "guoyww/animatediff-motion-lora-rolling-anticlockwise"),
+                    ("Roll right", "guoyww/animatediff-motion-lora-rolling-clockwise"),
+                ],
+                value="guoyww/animatediff-motion-lora-zoom-in",
+                interactive=True
+            )
+            
+            select_step = gr.Dropdown(
+                label='Inference steps',
+                choices=[
+                    ('1-Step', 1), 
+                    ('2-Step', 2),
+                    ('4-Step', 4),
+                    ('8-Step', 8),
+                ],
+                value=4,
+                interactive=True
+            )
+            
+            submit = gr.Button(
+                scale=1,
+                variant='primary'
+            )
+
+    # Output video display
+    video = gr.Video(
+        label='Generated Animation',
+        autoplay=True,
+        height=512,
+        width=512,
+        elem_id="video_output"
+    )
+
+    # Event handlers
+    gr.on(
+        triggers=[submit.click, prompt.submit],
+        fn=generate_image,
+        inputs=[prompt, select_base, select_motion, select_step],
+        outputs=[video],
+        api_name="instant_video",
+        queue=False
+    )
+
+    # Example prompts
+    gr.Examples(
+        examples=[
+            ["Focus: Eiffel Tower (Animate: Clouds moving)"],
+            ["Focus: Trees In forest (Animate: Lion running)"],
+            ["Focus: Astronaut in Space"],
+            ["Focus: Group of Birds in sky (Animate: Birds Moving) (Shot From distance)"],
+            ["Focus: Statue of liberty (Shot from Drone) (Animate: Drone coming toward statue)"],
+            ["Focus: Panda in Forest (Animate: Drinking Tea)"],
+            ["Focus: Kids Playing (Season: Winter)"],
+            ["Focus: Cars in Street (Season: Rain, Daytime) (Shot from Distance) (Movement: Cars running)"]
+        ],
+        fn=generate_image,
+        inputs=[prompt],
+        outputs=[video],
+        cache_examples="lazy",
+    )
+
+# Launch the interface
+demo.queue().launch()

requirements.txt

@@ -0,0 +1,8 @@
+gradio>=4.0.0
+torch
+diffusers
+transformers
+safetensors
+huggingface_hub
+Pillow
+spaces

style.css

@@ -0,0 +1,35 @@
+#video_output {
+    margin: auto;
+}
+
+.gradio-container {
+    font-family: 'Segoe UI', Tahoma, Geneva, Verdana, sans-serif;
+}
+
+h1 {
+    color: #2c3e50;
+    font-size: 2.5em;
+    margin-bottom: 1em;
+}
+
+.warning {
+    color: red;
+    font-weight: bold;
+}
+
+.examples-table {
+    margin-top: 2em;
+}
+
+.button-primary {
+    background-color: #3498db;
+    color: white;
+    padding: 10px 20px;
+    border-radius: 5px;
+    border: none;
+    cursor: pointer;
+}
+
+.button-primary:hover {
+    background-color: #2980b9;
+}