add decorator

rag_utils.py

@@ -9,6 +9,7 @@ from langchain.llms import HuggingFacePipeline
 from langchain.chat_models import ChatOpenAI
 from langchain.chains import RetrievalQA
 
+import spaces
 import config
 import torch
 print(torch.cuda.is_available())
@@ -23,6 +24,7 @@ _embedding_instance = None
 _model_instance = None
 _tokenizer = None
 
+@spaces.GPU
 def get_embedding_model():
     global _embedding_instance
     if _embedding_instance is None:
@@ -33,6 +35,7 @@ def get_embedding_model():
 
 # model_name = "Qwen/Qwen2.5-7B-Instruct-GPTQ-Int8" #"Qwen/Qwen2.5-7B-Instruct-AWQ" #"Qwen/Qwen2.5-7B-Instruct"
 
+@spaces.GPU
 def get_model():
     global _model_instance
     if _model_instance is None:
@@ -49,6 +52,7 @@ def get_model():
 
 # _model_instance.to(device)
 
+@spaces.GPU
 def get_tokenizer():
     global _tokenizer
     if _tokenizer is None: