updated the demo.yaml to support huggingface format

demo.yaml

@@ -15,6 +15,7 @@ backend:
   model_name: 
     openai: "gpt-3.5-turbo"
     azure: "azure/gpt-4"
+    huggingface: "huggingface/meta-llama/Meta-Llama-3-70B-Instruct"
   
   # ~~~ generation_parameters ~~
   n: 1
@@ -24,6 +25,7 @@ backend:
   top_p: 0.2
   frequency_penalty: 0
   presence_penalty: 0
+  stream: False
 
 n_api_retries: 6
 wait_time_between_retries: 20