update

Author: themrzmaster

llama_cpp/llama_chat_format.py

@@ -2462,7 +2462,7 @@ def base_function_calling(
         f"""root   ::= functions | "</done>"\n"""
         f"""functions ::= {function_names}\n"""
     )
-    
+
 
     prompt = template_renderer.render(
         messages=messages,
@@ -2517,8 +2517,7 @@ def base_function_calling(
                 mirostat_eta=mirostat_eta,
                 model=model,
                 logits_processor=logits_processor,
-            ),
-            stream=stream,
+            ),stream=stream
         )
 
     # One or more function calls
@@ -2804,16 +2803,16 @@ def vicuna_function_calling(
         "\nfunctions.{{ tool.function.name }}:\n"
         "{{ tool.function.parameters | tojson }}"
         "\n{% endfor %}"
-        "\n\nYou can respond to users messages with either a single message or multiple function calls."
-        "\n\nTo respond with a message begin the message with 'message:', use the following format:"
-        "\n\nmessage:"
-        "\n<message>"
+        "\n\nYou can respond to users messages with either a single message or multiple function calls, never both. If function calls are used, they must be the first part of the response."
         "\n\nTo respond with one or more function calls begin the message with 'functions.<function_name>:', use the following format:"
         "\n\nfunctions.<function_name>:"
         '\n{ "arg1": "value1", "arg2": "value2" };'
         "\nfunctions.<another_function_name>:"
         '\n{ "arg1": "value3", "arg2": "value4" }'
         "\n\nWhen you are done with the function calls, end the message with </done>."
+        "\n\nTo respond with a message begin the message with 'message:', use the following format:"
+        "\n\nmessage:"
+        "\n<message> </s>"
         "{% endif %}"
         "</s>\n"
         "{% endif %}"
@@ -2853,4 +2852,3 @@ def vicuna_function_calling(
     )
     return base_function_calling(end_token="</s>", 
                           **locals())
-