enable zerogpu

app.py

@@ -13,6 +13,7 @@ from transformers import AutoModelForCausalLM, AutoTokenizer, StoppingCriteria,
 from data import extract_leaves, split_document, handle_broken_output, clean_json_text, sync_empty_fields
 from examples import examples as input_examples
 from nuextract_logging import log_event
+import spaces
 
 
 MAX_INPUT_SIZE = 10_000
@@ -140,6 +141,7 @@ model = AutoModelForCausalLM.from_pretrained(model_name,
 tokenizer = AutoTokenizer.from_pretrained(model_name, use_auth_token=auth_token)
 model.eval()
 
+@spaces.GPU
 def gradio_interface_function(template, text, is_example):
     if len(tokenizer.tokenize(text)) > MAX_INPUT_SIZE:
         yield "", "Input text too long for space. Download model to use unrestricted.", ""

requirements.txt

@@ -1,3 +1,4 @@
 transformers
 torch
-accelerate
+accelerate
+spaces