Upload app.py

app.py

@@ -33,6 +33,7 @@ moondream.eval()
 #moondream.to_bettertransformer()
 
 #@spaces.GPU
+@torch.inference_mode()
 def answer_question(img, prompt):
     image_embeds = moondream.encode_image(img)
     streamer = TextIteratorStreamer(tokenizer, skip_special_tokens=True)