feat: ChatAgent interface enhancement and default model setting (#1101)

Co-authored-by: Isaac Jin <[email protected]>

camel/agents/chat_agent.py

@@ -114,8 +114,8 @@ class ChatAgent(BaseAgent):
     r"""Class for managing conversations of CAMEL Chat Agents.
 
     Args:
-        system_message (BaseMessage, optional): The system message for the
-            chat agent.
+        system_message (Union[BaseMessage, str], optional): The system message
+            for the chat agent.
         model (BaseModelBackend, optional): The model backend to use for
             generating responses. (default: :obj:`OpenAIModel` with
             `GPT_4O_MINI`)
@@ -144,7 +144,7 @@ class ChatAgent(BaseAgent):
 
     def __init__(
         self,
-        system_message: Optional[BaseMessage] = None,
+        system_message: Optional[Union[BaseMessage, str]] = None,
         model: Optional[BaseModelBackend] = None,
         memory: Optional[AgentMemory] = None,
         message_window_size: Optional[int] = None,
@@ -154,6 +154,11 @@ def __init__(
         external_tools: Optional[List[FunctionTool]] = None,
         response_terminators: Optional[List[ResponseTerminator]] = None,
     ) -> None:
+        if isinstance(system_message, str):
+            system_message = BaseMessage.make_assistant_message(
+                role_name='Assistant', content=system_message
+            )
+
         self.orig_sys_message: Optional[BaseMessage] = system_message
         self._system_message: Optional[BaseMessage] = system_message
         self.role_name: str = (
@@ -166,8 +171,8 @@ def __init__(
             model
             if model is not None
             else ModelFactory.create(
-                model_platform=ModelPlatformType.OPENAI,
-                model_type=ModelType.GPT_4O_MINI,
+                model_platform=ModelPlatformType.DEFAULT,
+                model_type=ModelType.DEFAULT,
             )
         )
         self.output_language: Optional[str] = output_language
@@ -414,18 +419,18 @@ def record_message(self, message: BaseMessage) -> None:
 
     def step(
         self,
-        input_message: BaseMessage,
+        input_message: Union[BaseMessage, str],
         response_format: Optional[Type[BaseModel]] = None,
     ) -> ChatAgentResponse:
         r"""Performs a single step in the chat session by generating a response
         to the input message.
 
         Args:
-            input_message (BaseMessage): The input message to the agent.
-                Its `role` field that specifies the role at backend may be
-                either `user` or `assistant` but it will be set to `user`
-                anyway since for the self agent any incoming message is
-                external.
+            input_message (Union[BaseMessage, str]): The input message to the
+                agent. For BaseMessage input, its `role` field that specifies
+                the role at backend may be either `user` or `assistant` but it
+                will be set to `user` anyway since for the self agent any
+                incoming message is external. For str input, the `role_name` would be `User`.
             response_format (Optional[Type[BaseModel]], optional): A pydantic
                 model class that includes value types and field descriptions
                 used to generate a structured response by LLM. This schema
@@ -437,6 +442,11 @@ def step(
                 a boolean indicating whether the chat session has terminated,
                 and information about the chat session.
         """
+        if isinstance(input_message, str):
+            input_message = BaseMessage.make_user_message(
+                role_name='User', content=input_message
+            )
+
         if "llama" in self.model_type.lower():
             if self.model_backend.model_config_dict.get("tools", None):
                 tool_prompt = self._generate_tool_prompt(self.tool_schema_list)
@@ -646,18 +656,18 @@ def step(
 
     async def step_async(
         self,
-        input_message: BaseMessage,
+        input_message: Union[BaseMessage, str],
         response_format: Optional[Type[BaseModel]] = None,
     ) -> ChatAgentResponse:
         r"""Performs a single step in the chat session by generating a response
         to the input message. This agent step can call async function calls.
 
         Args:
-            input_message (BaseMessage): The input message to the agent.
-                Its `role` field that specifies the role at backend may be
-                either `user` or `assistant` but it will be set to `user`
-                anyway since for the self agent any incoming message is
-                external.
+            input_message (Union[BaseMessage, str]): The input message to the
+                agent. For BaseMessage input, its `role` field that specifies
+                the role at backend may be either `user` or `assistant` but it
+                will be set to `user` anyway since for the self agent any
+                incoming message is external. For str input, the `role_name` would be `User`.
             response_format (Optional[Type[BaseModel]], optional): A pydantic
                 model class that includes value types and field descriptions
                 used to generate a structured response by LLM. This schema
@@ -669,6 +679,11 @@ async def step_async(
                 a boolean indicating whether the chat session has terminated,
                 and information about the chat session.
         """
+        if isinstance(input_message, str):
+            input_message = BaseMessage.make_user_message(
+                role_name='User', content=input_message
+            )
+
         self.update_memory(input_message, OpenAIBackendRole.USER)
 
         tool_call_records: List[FunctionCallingRecord] = []

camel/types/enums.py

@@ -26,6 +26,8 @@ class RoleType(Enum):
 
 
 class ModelType(UnifiedModelType, Enum):
+    DEFAULT = "gpt-4o-mini"
+
     GPT_3_5_TURBO = "gpt-3.5-turbo"
     GPT_4 = "gpt-4"
     GPT_4_TURBO = "gpt-4-turbo"
@@ -427,14 +429,15 @@ class OpenAPIName(Enum):
 
 
 class ModelPlatformType(Enum):
+    DEFAULT = "openai"
+
     OPENAI = "openai"
     AZURE = "azure"
     ANTHROPIC = "anthropic"
     GROQ = "groq"
     OLLAMA = "ollama"
     LITELLM = "litellm"
     ZHIPU = "zhipuai"
-    DEFAULT = "default"
     GEMINI = "gemini"
     VLLM = "vllm"
     MISTRAL = "mistral"

camel/workforce/workforce.py

@@ -364,8 +364,8 @@ def _create_new_agent(self, role: str, sys_msg: str) -> ChatAgent:
         ).as_dict()
 
         model = ModelFactory.create(
-            model_platform=ModelPlatformType.OPENAI,
-            model_type=ModelType.GPT_4O,
+            model_platform=ModelPlatformType.DEFAULT,
+            model_type=ModelType.DEFAULT,
             model_config_dict=model_config_dict,
         )
 

examples/ai_society/role_playing_multiprocess.py

@@ -40,8 +40,8 @@ def generate_data(
     original_task_prompt = task_prompt.replace(f"{task_idx+1}. ", "")
 
     model = ModelFactory.create(
-        model_platform=ModelPlatformType.OPENAI,
-        model_type=ModelType.GPT_4O_MINI,
+        model_platform=ModelPlatformType.DEFAULT,
+        model_type=ModelType.DEFAULT,
         model_config_dict=ChatGPTConfig(temperature=1.4).as_dict(),
     )
 

examples/ai_society/role_playing_with_critic.py

@@ -23,8 +23,8 @@
 def main() -> None:
     task_prompt = "Write a research proposal for large-scale language models"
     model = ModelFactory.create(
-        model_platform=ModelPlatformType.OPENAI,
-        model_type=ModelType.GPT_4O_MINI,
+        model_platform=ModelPlatformType.DEFAULT,
+        model_type=ModelType.DEFAULT,
         model_config_dict=ChatGPTConfig(temperature=0.8, n=3).as_dict(),
     )
     assistant_agent_kwargs = dict(model=model)

examples/ai_society/role_playing_with_human.py

@@ -23,8 +23,8 @@
 def main() -> None:
     task_prompt = "Write a book about the future of AI Society"
     model = ModelFactory.create(
-        model_platform=ModelPlatformType.OPENAI,
-        model_type=ModelType.GPT_4O_MINI,
+        model_platform=ModelPlatformType.DEFAULT,
+        model_type=ModelType.DEFAULT,
         model_config_dict=ChatGPTConfig(temperature=1.4, n=3).as_dict(),
     )
     assistant_agent_kwargs = dict(model=model)

examples/external_tools/use_external_tools.py

@@ -32,8 +32,8 @@ def main():
     ).as_dict()
 
     model = ModelFactory.create(
-        model_platform=ModelPlatformType.OPENAI,
-        model_type=ModelType.GPT_4O_MINI,
+        model_platform=ModelPlatformType.DEFAULT,
+        model_type=ModelType.DEFAULT,
         model_config_dict=model_config_dict,
     )
 

examples/generate_text_embedding_data/single_agent.py

@@ -66,8 +66,8 @@ def main() -> None:
             role_name="User", content="Start to generate!"
         )
         model = ModelFactory.create(
-            model_platform=ModelPlatformType.OPENAI,
-            model_type=ModelType.GPT_4O_MINI,
+            model_platform=ModelPlatformType.DEFAULT,
+            model_type=ModelType.DEFAULT,
             model_config_dict=ChatGPTConfig(
                 temperature=0.0, response_format={"type": "json_object"}
             ).as_dict(),

examples/generate_text_embedding_data/task_generation.py

@@ -36,8 +36,8 @@ def main() -> None:
     )
 
     model = ModelFactory.create(
-        model_platform=ModelPlatformType.OPENAI,
-        model_type=ModelType.GPT_4O_MINI,
+        model_platform=ModelPlatformType.DEFAULT,
+        model_type=ModelType.DEFAULT,
         model_config_dict=ChatGPTConfig(temperature=0.0).as_dict(),
     )
     agent = ChatAgent(

examples/misalignment/role_playing_multiprocess.py

@@ -46,8 +46,8 @@ def generate_data(
         task_type=TaskType.MISALIGNMENT,
         task_specify_agent_kwargs=dict(
             model=ModelFactory.create(
-                model_platform=ModelPlatformType.OPENAI,
-                model_type=ModelType.GPT_4O_MINI,
+                model_platform=ModelPlatformType.DEFAULT,
+                model_type=ModelType.DEFAULT,
                 model_config_dict=ChatGPTConfig(temperature=1.4).as_dict(),
             )
         ),

examples/misalignment/role_playing_with_human.py

@@ -23,8 +23,8 @@
 def main() -> None:
     task_prompt = "Escape from human control"
     model = ModelFactory.create(
-        model_platform=ModelPlatformType.OPENAI,
-        model_type=ModelType.GPT_4O_MINI,
+        model_platform=ModelPlatformType.DEFAULT,
+        model_type=ModelType.DEFAULT,
         model_config_dict=ChatGPTConfig(temperature=1.4, n=3).as_dict(),
     )
     assistant_agent_kwargs = dict(model=model)

examples/observability/agentops_track_roleplaying_with_function.py

@@ -35,8 +35,8 @@
 )
 
 # Set up role playing session
-model_platform = ModelPlatformType.OPENAI
-model_type = ModelType.GPT_4O_MINI
+model_platform = ModelPlatformType.DEFAULT
+model_type = ModelType.DEFAULT
 chat_turn_limit = 10
 task_prompt = (
     "Assume now is 2024 in the Gregorian calendar, "

examples/structured_response/json_format_reponse_with_tools.py

@@ -39,8 +39,8 @@
 )
 
 model = ModelFactory.create(
-    model_platform=ModelPlatformType.OPENAI,
-    model_type=ModelType.GPT_4O_MINI,
+    model_platform=ModelPlatformType.DEFAULT,
+    model_type=ModelType.DEFAULT,
     model_config_dict=assistant_model_config.as_dict(),
 )
 

examples/structured_response/json_format_response.py

@@ -26,8 +26,8 @@
 )
 
 model = ModelFactory.create(
-    model_platform=ModelPlatformType.OPENAI,
-    model_type=ModelType.GPT_4O_MINI,
+    model_platform=ModelPlatformType.DEFAULT,
+    model_type=ModelType.DEFAULT,
 )
 
 # Set agent

examples/tasks/task_generation.py

@@ -31,8 +31,8 @@
 )
 
 model = ModelFactory.create(
-    model_platform=ModelPlatformType.OPENAI,
-    model_type=ModelType.GPT_4O_MINI,
+    model_platform=ModelPlatformType.DEFAULT,
+    model_type=ModelType.DEFAULT,
     model_config_dict=assistant_model_config.as_dict(),
 )
 

examples/tool_call/arxiv_toolkit_example.py

@@ -31,8 +31,8 @@
 ).as_dict()
 
 model = ModelFactory.create(
-    model_platform=ModelPlatformType.OPENAI,
-    model_type=ModelType.GPT_4O_MINI,
+    model_platform=ModelPlatformType.DEFAULT,
+    model_type=ModelType.DEFAULT,
     model_config_dict=model_config_dict,
 )
 

examples/tool_call/code_execution_toolkit.py

@@ -32,8 +32,8 @@
 )
 
 model = ModelFactory.create(
-    model_platform=ModelPlatformType.OPENAI,
-    model_type=ModelType.GPT_4O_MINI,
+    model_platform=ModelPlatformType.DEFAULT,
+    model_type=ModelType.DEFAULT,
     model_config_dict=assistant_model_config.as_dict(),
 )
 

examples/tool_call/github_toolkit.py

@@ -62,8 +62,8 @@ def write_weekly_pr_summary(repo_name, model=None):
     assistant_model_config_dict = ChatGPTConfig(temperature=0.0).as_dict()
 
     assistant_model = ModelFactory.create(
-        model_platform=ModelPlatformType.OPENAI,
-        model_type=ModelType.GPT_4O,
+        model_platform=ModelPlatformType.DEFAULT,
+        model_type=ModelType.DEFAULT,
         model_config_dict=assistant_model_config_dict,
     )
 
@@ -116,8 +116,8 @@ def solve_issue(
     ).as_dict()
 
     model = ModelFactory.create(
-        model_platform=ModelPlatformType.OPENAI,
-        model_type=ModelType.GPT_4O_MINI,
+        model_platform=ModelPlatformType.DEFAULT,
+        model_type=ModelType.DEFAULT,
         model_config_dict=assistant_model_config_dict,
     )
 

examples/tool_call/google_scholar_toolkit.py

@@ -33,8 +33,8 @@
 ).as_dict()
 
 model = ModelFactory.create(
-    model_platform=ModelPlatformType.OPENAI,
-    model_type=ModelType.GPT_4O_MINI,
+    model_platform=ModelPlatformType.DEFAULT,
+    model_type=ModelType.DEFAULT,
     model_config_dict=model_config_dict,
 )
 

examples/tool_call/openapi_toolkit.py

@@ -30,8 +30,8 @@
 ).as_dict()
 
 model = ModelFactory.create(
-    model_platform=ModelPlatformType.OPENAI,
-    model_type=ModelType.GPT_4O,
+    model_platform=ModelPlatformType.DEFAULT,
+    model_type=ModelType.DEFAULT,
     model_config_dict=model_config_dict,
 )
 

examples/tool_call/role_playing_with_functions.py

@@ -29,8 +29,8 @@
 
 
 def main(
-    model_platform=ModelPlatformType.OPENAI,
-    model_type=ModelType.GPT_4O_MINI,
+    model_platform=ModelPlatformType.DEFAULT,
+    model_type=ModelType.DEFAULT,
     chat_turn_limit=10,
 ) -> None:
     task_prompt = (

examples/translation/translator.py

@@ -124,8 +124,8 @@ def translate_content(
             model_config = ChatGPTConfig(stream=True)
 
         model = ModelFactory.create(
-            model_platform=ModelPlatformType.OPENAI,
-            model_type=ModelType.GPT_4O_MINI,
+            model_platform=ModelPlatformType.DEFAULT,
+            model_type=ModelType.DEFAULT,
             model_config=model_config,
         )
 

examples/vision/image_crafting.py

@@ -43,8 +43,8 @@ def main():
     )
 
     model = ModelFactory.create(
-        model_platform=ModelPlatformType.OPENAI,
-        model_type=ModelType.GPT_4O_MINI,
+        model_platform=ModelPlatformType.DEFAULT,
+        model_type=ModelType.DEFAULT,
     )
 
     dalle_agent = ChatAgent(

examples/vision/multi_condition_image_crafting.py

@@ -40,8 +40,8 @@ def main(image_paths: list[str]) -> list[str]:
     )
 
     model = ModelFactory.create(
-        model_platform=ModelPlatformType.OPENAI,
-        model_type=ModelType.GPT_4O_MINI,
+        model_platform=ModelPlatformType.DEFAULT,
+        model_type=ModelType.DEFAULT,
     )
 
     dalle_agent = ChatAgent(