Update model_llm.py

model_llm.py

@@ -15,13 +15,19 @@ from transformers import BitsAndBytesConfig
 from threading import Thread
 
 model_name = "mistralai/Mistral-7B-Instruct-v0.3"
+auth_token = os.environ.get("HF_TOKEN")
 
 # Auto-regressive model for language completion: padding left
-tokenizer = AutoTokenizer.from_pretrained(model_name, padding_side="left")
+tokenizer = AutoTokenizer.from_pretrained(
+    model_name,
+    padding_side="left",
+    use_auth_token=auth_token
+)
 model = AutoModelForCausalLM.from_pretrained(
     model_name,
     device_map="auto",
     torch_dtype=torch.float16,
-    low_cpu_mem_usage=True
+    low_cpu_mem_usage=True,
+    use_auth_token=auth_token
 )
 model = torch.compile(model)