Update app.py

app.py

@@ -1,5 +1,6 @@
 import gradio as gr
 import os
+import spaces
 import torch
 from transformers import AutoTokenizer, AutoModelForCausalLM
 
@@ -7,6 +8,7 @@ model_path = "cognitivecomputations/dolphin-2.8-mistral-7b-v02"
 tokenizer = AutoTokenizer.from_pretrained(model_path)
 model = AutoModelForCausalLM.from_pretrained(model_path)
 
+@spaces.GPU
 def chat(prompt):
     input_ids = tokenizer.encode(prompt, return_tensors="pt")
     output = model.generate(input_ids, max_length=1024, num_return_sequences=1, top_p=0.9, top_k=50, num_beams=2, early_stopping=True)