update to bf16 model

app.py

@@ -254,7 +254,7 @@ def generate_response(message, history, prompt_choice):
     
     full_prompt = base_prompt + "\n" + formatted_input
     completion = client.completions.create(
-        model="meta-llama/Meta-Llama-3.1-405B-FP8",
+        model="meta-llama/Meta-Llama-3.1-405B",
         prompt=full_prompt,
         temperature=0.7,
         frequency_penalty=0.1,