Create app.py

app.py

@@ -0,0 +1,36 @@
+import gradio as gr
+from transformers import AutoProcessor, LlavaForConditionalGeneration
+from PIL import Image
+import requests
+from io import BytesIO
+
+# Load the Pixtral model and processor
+model_id = "mistral-community/pixtral-12b"
+processor = AutoProcessor.from_pretrained(model_id)
+model = LlavaForConditionalGeneration.from_pretrained(model_id).to("cuda")
+
+def generate_text(input_text, image_url):
+    # Load image
+    response = requests.get(image_url)
+    image = Image.open(BytesIO(response.content))
+
+    # Prepare inputs
+    inputs = processor(text=input_text, images=image, return_tensors="pt").to("cuda")
+
+    # Generate output
+    outputs = model.generate(**inputs)
+    generated_text = processor.batch_decode(outputs, skip_special_tokens=True)[0]
+
+    return generated_text
+
+# Create Gradio interface
+iface = gr.Interface(
+    fn=generate_text,
+    inputs=[gr.Textbox(label="Enter your text here"), gr.Textbox(label="Enter image URL")],
+    outputs=gr.Textbox(label="Generated Text"),
+    title="Pixtral Model Interaction",
+    description="Interact with the Pixtral model using text and image inputs."
+)
+
+# Launch the interface
+iface.launch()