Add zero gpu support

app.py

@@ -5,6 +5,7 @@ from transformers import MistralForCausalLM, TextIteratorStreamer, AutoTokenizer
 from time import sleep
 from threading import Thread
 from torch import float16
+import spaces
 
 config = PeftConfig.from_pretrained("lang-uk/dragoman")
 quant_config = BitsAndBytesConfig(
@@ -20,6 +21,7 @@ model = MistralForCausalLM.from_pretrained("mistralai/Mistral-7B-v0.1",
 model = PeftModel.from_pretrained(model, "lang-uk/dragoman")
 tokenizer = AutoTokenizer.from_pretrained("mistralai/Mistral-7B-v0.1", use_fast=False, add_bos_token=False)
 
+@spaces.GPU(duration=200)
 def translate(input_text):
     # iteratively generate
     input_text = input_text.strip()