import spaces for zerogpu

app.py

@@ -1,3 +1,4 @@
+import spaces
 import gradio as gr
 import torch
 from transformers import AutoModelForCausalLM, AutoTokenizer, StoppingCriteria, StoppingCriteriaList, TextIteratorStreamer
@@ -37,6 +38,8 @@ Your answer should be friendly, unbiased, faithful, informative and detailed.'
 system_prompt = f"<|im_start|>{system_role}\n{system_prompt}<|im_end|>"
 
 # Function to generate model predictions.
+
+@spaces.GPU()
 def predict(message, history):
     # history = []
     history_transformer_format = history + [[message, ""]]