Create app.py

app.py

@@ -0,0 +1,46 @@
+import torch
+from PIL import Image
+from transformers import AutoModel, CLIPImageProcessor
+import gradio as gr
+
+# Load the model
+model = AutoModel.from_pretrained(
+    'OpenGVLab/InternVL2_5-1B',
+    torch_dtype=torch.float32,  # Use float32 for CPU compatibility
+    low_cpu_mem_usage=True,
+    trust_remote_code=True,
+    use_flash_attn=False  # Disable Flash Attention
+).eval()  # Do not move to CUDA, force CPU execution
+
+# Load the image processor
+image_processor = CLIPImageProcessor.from_pretrained('OpenGVLab/InternVL2_5-1B')
+
+# Define the function to process the image and generate outputs
+def process_image(image):
+    try:
+        # Convert uploaded image to RGB
+        image = image.convert('RGB')
+        
+        # Preprocess the image
+        pixel_values = image_processor(images=image, return_tensors='pt').pixel_values
+        
+        # Run the model on CPU
+        outputs = model(pixel_values)
+        
+        # Assuming the model returns embeddings or features
+        return f"Output Shape: {outputs.last_hidden_state.shape}"
+    except Exception as e:
+        return f"Error: {str(e)}"
+
+# Create the Gradio interface
+demo = gr.Interface(
+    fn=process_image,  # Function to process the input
+    inputs=gr.Image(type="pil"),  # Accepts images as input
+    outputs=gr.Textbox(label="Model Output"),  # Displays model output
+    title="InternViT Demo",
+    description="Upload an image to process it using the InternViT model from OpenGVLab."
+)
+
+# Launch the demo
+if __name__ == "__main__":
+    demo.launch(server_name="0.0.0.0", server_port=7860)