Update app.py

app.py

@@ -27,12 +27,14 @@ text_generator = pipeline("text-generation", model=model_gpt2, tokenizer=tokeniz
 
 # Load the Llama model and tokenizer at the beginning
 model_name = "meta-llama/Meta-Llama-3.1-8B"
+tokenizer_llama = AutoTokenizer.from_pretrained(model_name, token=hf_token)
 model_llama = AutoModelForCausalLM.from_pretrained(
     model_name,
-    torch_dtype=torch.float16,  # Use FP16 for reduced memory
+    torch_dtype="auto",
+    device_map="auto",
     token=hf_token
 )
-tokenizer_llama = AutoTokenizer.from_pretrained(model_name, token=hf_token)
+