Update app.py

app.py

@@ -52,13 +52,14 @@ def describe_image(image_array):
     inputs = processor(text=prompt, images=[image], return_tensors="pt")
 
     # Inference: Generation of the output
-    generated_ids = model.generate(
-        **inputs, 
-        max_new_tokens=500,
-        num_beams=1,  # Disable beam search 
-        do_sample=False,  # Disable sampling
-        temperature=1.0  # Set temperature to 1.0
-    )
+    with torch.no_grad():
+        generated_ids = model.generate(
+            **inputs, 
+            max_new_tokens=500,
+            num_beams=1,  # Disable beam search 
+            do_sample=False,  # Disable sampling
+            temperature=1.0  # Set temperature to 1.0
+        )
     output_ids = [
         generated_ids[len(input_ids) :]
         for input_ids, generated_ids in zip(inputs.input_ids, generated_ids)
@@ -74,8 +75,8 @@ iface = gr.Interface(
     fn=describe_image,
     inputs=gr.Image(),
     outputs=[gr.Textbox(label="Description"), gr.JSON(label="Embeddings")],
-    title="Image Description with Qwen Model",
-    description="Upload an image to get a detailed description using the Qwen2-VL-7B-Instruct model."
+    title="Image Description with SmolVLM Model",
+    description="Upload an image to get a detailed description using the SmolVLM-500M-Instruct model."
 )
 
 # Launch the app