Update app.py

app.py

@@ -2,10 +2,13 @@ import gradio as gr
 import re
 from transformers import AutoModelForCausalLM, AutoTokenizer
 
-model_name_or_path = "TheBloke/OpenHermes-2-Mistral-7B-AWQ"
-model = AutoAWQForCausalLM.from_quantized(model_name_or_path, fuse_layers=True,
-                                          trust_remote_code=False, safetensors=True)
-tokenizer = AutoTokenizer.from_pretrained(model_name_or_path, trust_remote_code=False)
+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_4bit=True,
+                                             revision="main")
+tokenizer = AutoTokenizer.from_pretrained(model_name_or_path, use_fast=True)
 
 BASE_SYSTEM_MESSAGE = "I carefully provide accurate, factual, thoughtful, nuanced answers and am brilliant at reasoning."