Update app.py

app.py

@@ -11,6 +11,8 @@ model_name_or_path = "teknium/OpenHermes-2-Mistral-7B"
 model = AutoModelForCausalLM.from_pretrained(model_name_or_path,
                                              device_map="auto",
                                              trust_remote_code=False,
+                                             load_in_8bit=True,
+                                             use_flash_attention_2=True,
                                              revision="main")
 
 tokenizer = AutoTokenizer.from_pretrained(model_name_or_path, use_fast=True)