Update app.py

app.py

@@ -1,5 +1,6 @@
 import gradio as gr
 from huggingface_hub import InferenceClient
+import spaces
 
 """
 For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
@@ -7,6 +8,7 @@ For more information on `huggingface_hub` Inference API support, please check th
 client = InferenceClient("MakiAi/Llama-3.2-3B-Instruct-bnb-4bit-OKU_wiki_llama3.1_8b_inst_Reflexive_chunk200_overlap700-10epochs")
 
 
+@spaces.GPU
 def respond(
     message,
     history: list[tuple[str, str]],