Update app.py

app.py

@@ -19,9 +19,6 @@ from langchain.prompts import PromptTemplate
 from langchain.prompts.prompt import PromptTemplate
 from langchain.chat_models import ChatOpenAI
 
-LOAD_MODEL=False
-tokenizer,model = None,None
-
 def load_model():
     tokenizer = AutoTokenizer.from_pretrained("THUDM/chatglm-6b", trust_remote_code=True)
     model = AutoModel.from_pretrained("THUDM/chatglm-6b-int4", trust_remote_code=True).quantize(bits=4, compile_parallel_kernel=True, parallel_num=2).float()
@@ -32,9 +29,8 @@ def load_model():
 def chat_glm(input, history=None):
     if history is None:
         history = []
-    if not LOAD_MODEL:
-        LOAD_MODEL=True
-        tokenizer,model = load_model()
+
+    tokenizer,model = load_model()
     response, history = model.chat(tokenizer, input, history)
     logger.info("chatglm:", input,response)
     return history, history