Merge ae3370db31 into 5ad267707d

autogen/agentchat/conversable_agent.py

@@ -257,17 +257,36 @@ class ConversableAgent(LLMAgent):
         }
 
     def _validate_llm_config(self, llm_config):
-        assert llm_config in (None, False) or isinstance(
-            llm_config, dict
-        ), "llm_config must be a dict or False or None."
+        assert llm_config in (None, False) or isinstance(llm_config, dict), "llm_config must be a dict, False, or None."
+
         if llm_config is None:
             llm_config = self.DEFAULT_CONFIG
-        self.llm_config = self.DEFAULT_CONFIG if llm_config is None else llm_config
-        # TODO: more complete validity check
-        if self.llm_config in [{}, {"config_list": []}, {"config_list": [{"model": ""}]}]:
+
+        self.llm_config = llm_config
+
+        if isinstance(self.llm_config, dict):
+            config_list = self.llm_config.get("config_list", [])
+            if not isinstance(config_list, list):
+                raise ValueError("llm_config: 'config_list' must be a list.")
+
+            # If config_list is empty, check if 'model' field is present in llm_config
+            if not config_list and "model" not in self.llm_config:
+                raise ValueError("llm_config: 'model' field is required in 'llm_config' or each 'config_list' entry.")
+
+            # When config_list is not empty, check each item in config_list
+            for config in config_list:
+                if not isinstance(config, dict):
+                    raise ValueError("llm_config: 'config_list' must be a list of dictionaries.")
+                if "model" not in config or not config["model"]:
                     raise ValueError(
-                "When using OpenAI or Azure OpenAI endpoints, specify a non-empty 'model' either in 'llm_config' or in each config of 'config_list'."
+                        "llm_config: 'model' field is required for each item in 'config_list' and must not be empty."
                     )
+                if "api_key" in config and not isinstance(config["api_key"], str):
+                    raise ValueError("llm_config: 'api_key' must be a string.")
+                if "tags" in config:
+                    if not isinstance(config["tags"], list) or not all(isinstance(tag, str) for tag in config["tags"]):
+                        raise ValueError("llm_config: 'tags' must be a list of strings.")
+
         self.client = None if self.llm_config is False else OpenAIWrapper(**self.llm_config)
 
     @staticmethod

test/agentchat/test_conversable_agent.py

@@ -831,10 +831,28 @@ def test_register_for_llm_without_LLM():
             return f"{s} done"
 
 
+def test_register_for_llm_with_valid_configuration():
+    with pytest.MonkeyPatch.context() as mp:
+        mp.setenv("OPENAI_API_KEY", MOCK_OPEN_AI_API_KEY)
+
+        valid_config = {
+            "config_list": [
+                {"model": "gpt-4"},
+                {"model": "gpt-4", "api_key": MOCK_OPEN_AI_API_KEY, "tags": ["gpt4", "openai"]},
+            ],
+        }
+        assistant = ConversableAgent(
+            name="assistant",
+            llm_config=valid_config,
+        )
+        assert assistant is not None
+        assert assistant.llm_config == valid_config
+
+
 def test_register_for_llm_without_configuration():
     with pytest.raises(
         ValueError,
-        match="When using OpenAI or Azure OpenAI endpoints, specify a non-empty 'model' either in 'llm_config' or in each config of 'config_list'.",
+        match="llm_config: 'model' field is required in 'llm_config' or each 'config_list' entry.",
     ):
         ConversableAgent(name="agent", llm_config={"config_list": []})
 
@@ -842,11 +860,27 @@ def test_register_for_llm_without_configuration():
 def test_register_for_llm_without_model_name():
     with pytest.raises(
         ValueError,
-        match="When using OpenAI or Azure OpenAI endpoints, specify a non-empty 'model' either in 'llm_config' or in each config of 'config_list'.",
+        match="llm_config: 'model' field is required for each item in 'config_list' and must not be empty.",
     ):
         ConversableAgent(name="agent", llm_config={"config_list": [{"model": ""}]})
 
 
+def test_register_for_llm_with_invalid_tags():
+    with pytest.raises(
+        ValueError,
+        match="llm_config: 'tags' must be a list of strings.",
+    ):
+        ConversableAgent(name="agent", llm_config={"config_list": [{"model": "gpt-4", "tags": "invalid_tags"}]})
+
+
+def test_register_for_llm_with_invalid_api_key():
+    with pytest.raises(
+        ValueError,
+        match="llm_config: 'api_key' must be a string.",
+    ):
+        ConversableAgent(name="agent", llm_config={"config_list": [{"model": "gpt-4", "api_key": 1234}]})
+
+
 def test_register_for_execution():
     with pytest.MonkeyPatch.context() as mp:
         mp.setenv("OPENAI_API_KEY", MOCK_OPEN_AI_API_KEY)
@@ -1606,7 +1640,7 @@ def test_http_client():
 
 def test_adding_duplicate_function_warning():
 
-    config_base = [{"base_url": "http://0.0.0.0:8000", "api_key": "NULL"}]
+    config_base = [{"base_url": "http://0.0.0.0:8000", "api_key": "NULL", "model": "na"}]
 
     agent = autogen.ConversableAgent(
         "jtoy",