Update app.py

app.py

@@ -1,10 +1,12 @@
 import gradio as gr
-# Load model directly
+import torch
+import spaces
 from transformers import AutoTokenizer, AutoModelForCausalLM
 
 tokenizer = AutoTokenizer.from_pretrained("ping98k/typhoon-7b-rag-instruct-th")
 model = AutoModelForCausalLM.from_pretrained("ping98k/typhoon-7b-rag-instruct-th")
 
+@spaces.GPU(duration=120)
 def response(instruction, history, inputText):
     inp = f"""### Instruction:
 {instruction}