Additional updates from 3.5 to 4o-mini and 4 to 4o

Signed-off-by: Mark Sze <[email protected]>

Author: marklysze

test/agentchat/contrib/agent_eval/test_agent_eval.py

@@ -38,13 +38,12 @@ def remove_ground_truth(test_case: str):
         filter_dict={
             "api_type": ["openai"],
             "model": [
+                "gpt-4o-mini",
+                "gpt-4o",
                 "gpt-4-turbo",
                 "gpt-4-turbo-preview",
                 "gpt-4-0125-preview",
                 "gpt-4-1106-preview",
-                "gpt-3.5-turbo",
-                "gpt-3.5-turbo-0125",
-                "gpt-3.5-turbo-1106",
             ],
         },
     )

test/agentchat/contrib/capabilities/chat_with_teachable_agent.py

@@ -17,12 +17,8 @@
 from test_assistant_agent import KEY_LOC, OAI_CONFIG_LIST  # noqa: E402
 
 # Specify the model to use. GPT-3.5 is less reliable than GPT-4 at learning from user input.
-filter_dict = {"model": ["gpt-4-0125-preview"]}
-# filter_dict = {"model": ["gpt-3.5-turbo-1106"]}
-# filter_dict = {"model": ["gpt-4-0613"]}
-# filter_dict = {"model": ["gpt-3.5-turbo"]}
-# filter_dict = {"model": ["gpt-4"]}
-# filter_dict = {"model": ["gpt-35-turbo-16k", "gpt-3.5-turbo-16k"]}
+filter_dict = {"model": ["gpt-4o-mini"]}
+# filter_dict = {"model": ["gpt-4-0125-preview"]}
 
 
 def create_teachable_agent(reset_db=False):

test/agentchat/contrib/capabilities/test_image_generation_capability.py

@@ -32,7 +32,7 @@
 sys.path.append(os.path.join(os.path.dirname(__file__), "../.."))
 from conftest import MOCK_OPEN_AI_API_KEY, skip_openai  # noqa: E402
 
-filter_dict = {"model": ["gpt-35-turbo-16k", "gpt-3.5-turbo-16k"]}
+filter_dict = {"model": ["gpt-4o-mini"]}
 
 RESOLUTIONS = ["256x256", "512x512", "1024x1024"]
 QUALITIES = ["standard", "hd"]
@@ -67,21 +67,21 @@ def api_key():
 
 @pytest.fixture
 def dalle_config() -> Dict[str, Any]:
-    config_list = openai_utils.config_list_from_models(model_list=["dall-e-2"], exclude="aoai")
+    config_list = openai_utils.config_list_from_models(model_list=["dall-e-3"], exclude="aoai")
     if not config_list:
-        config_list = [{"model": "dall-e-2", "api_key": api_key()}]
+        config_list = [{"model": "dall-e-3", "api_key": api_key()}]
     return {"config_list": config_list, "timeout": 120, "cache_seed": None}
 
 
 @pytest.fixture
-def gpt3_config() -> Dict[str, Any]:
+def gpt4_config() -> Dict[str, Any]:
     config_list = [
         {
-            "model": "gpt-35-turbo-16k",
+            "model": "gpt-4o-mini",
             "api_key": api_key(),
         },
         {
-            "model": "gpt-3.5-turbo-16k",
+            "model": "gpt-4o",
             "api_key": api_key(),
         },
     ]

test/agentchat/contrib/capabilities/test_teachable_agent.py

@@ -31,10 +31,8 @@
 # Specify the model to use by uncommenting one of the following lines.
 # filter_dict={"model": ["gpt-4-1106-preview"]}
 # filter_dict={"model": ["gpt-4-0613"]}
-# filter_dict={"model": ["gpt-3.5-turbo-1106"]}
-# filter_dict={"model": ["gpt-3.5-turbo-0613"]}
 # filter_dict={"model": ["gpt-4"]}
-filter_dict = {"tags": ["gpt-35-turbo-16k", "gpt-3.5-turbo-16k"]}
+filter_dict = {"tags": ["gpt-4o-mini"]}
 
 
 def create_teachable_agent(reset_db=False, verbosity=0):

test/agentchat/contrib/capabilities/test_transform_messages.py

@@ -33,7 +33,7 @@ def test_transform_messages_capability():
             OAI_CONFIG_LIST,
             KEY_LOC,
             filter_dict={
-                "model": "gpt-3.5-turbo",
+                "model": "gpt-4o-mini",
             },
         )
 

test/agentchat/contrib/test_agent_builder.py

@@ -51,8 +51,8 @@ def test_build():
     builder = AgentBuilder(
         config_file_or_env=OAI_CONFIG_LIST,
         config_file_location=KEY_LOC,
-        builder_model_tags=["gpt-4"],
-        agent_model_tags=["gpt-4"],
+        builder_model_tags=["gpt-4o"],
+        agent_model_tags=["gpt-4o"],
     )
     building_task = (
         "Find a paper on arxiv by programming, and analyze its application in some domain. "
@@ -83,8 +83,8 @@ def test_build_from_library():
     builder = AgentBuilder(
         config_file_or_env=OAI_CONFIG_LIST,
         config_file_location=KEY_LOC,
-        builder_model_tags=["gpt-4"],
-        agent_model_tags=["gpt-4"],
+        builder_model_tags=["gpt-4o"],
+        agent_model_tags=["gpt-4o"],
     )
     building_task = (
         "Find a paper on arxiv by programming, and analyze its application in some domain. "
@@ -136,8 +136,8 @@ def test_save():
     builder = AgentBuilder(
         config_file_or_env=OAI_CONFIG_LIST,
         config_file_location=KEY_LOC,
-        builder_model_tags=["gpt-4"],
-        agent_model_tags=["gpt-4"],
+        builder_model_tags=["gpt-4o"],
+        agent_model_tags=["gpt-4o"],
     )
     building_task = (
         "Find a paper on arxiv by programming, and analyze its application in some domain. "
@@ -175,8 +175,8 @@ def test_load():
         config_file_location=KEY_LOC,
         # builder_model=["gpt-4", "gpt-4-1106-preview"],
         # agent_model=["gpt-4", "gpt-4-1106-preview"],
-        builder_model_tags=["gpt-4"],
-        agent_model_tags=["gpt-4"],
+        builder_model_tags=["gpt-4o"],
+        agent_model_tags=["gpt-4o"],
     )
 
     config_save_path = f"{here}/example_test_agent_builder_config.json"
@@ -204,8 +204,8 @@ def test_clear_agent():
     builder = AgentBuilder(
         config_file_or_env=OAI_CONFIG_LIST,
         config_file_location=KEY_LOC,
-        builder_model_tags=["gpt-4"],
-        agent_model_tags=["gpt-4"],
+        builder_model_tags=["gpt-4o"],
+        agent_model_tags=["gpt-4o"],
     )
 
     config_save_path = f"{here}/example_test_agent_builder_config.json"

test/agentchat/contrib/test_gpt_assistant.py

@@ -40,9 +40,6 @@
                 "gpt-4-turbo-preview",
                 "gpt-4-0125-preview",
                 "gpt-4-1106-preview",
-                "gpt-3.5-turbo",
-                "gpt-3.5-turbo-0125",
-                "gpt-3.5-turbo-1106",
             ],
         },
     )

test/agentchat/contrib/test_llamaindex_conversable_agent.py

@@ -48,7 +48,7 @@ def test_group_chat_with_llama_index_conversable_agent(chat_mock: MagicMock) ->
     Each agent is set to describe an image in a unique style, but the chat should not exceed the specified max_rounds.
     """
     llm = OpenAI(
-        model="gpt-4",
+        model="gpt-4o",
         temperature=0.0,
         api_key=openaiKey,
     )

test/agentchat/contrib/test_reasoning_agent.py

@@ -43,7 +43,7 @@ def think_node():
 @pytest.fixture
 def reasoning_agent():
     """Create a ReasoningAgent instance for testing"""
-    config_list = [{"model": "gpt-4", "api_key": "fake_key"}]
+    config_list = [{"model": "gpt-4o", "api_key": "fake_key"}]
     llm_config = {"config_list": config_list, "temperature": 0}
     return ReasoningAgent("reasoning_agent", llm_config=llm_config)
 
@@ -164,7 +164,10 @@ def test_reasoning_agent_answer():
 
 def helper_test_reasoning_agent_answer(max_depth, beam_size, answer_approach):
     """Test that ReasoningAgent properly terminates when TERMINATE is received"""
-    mock_config = {"config_list": [{"model": "gpt-4", "api_key": "fake", "base_url": "0.0.0.0:8000"}], "temperature": 0}
+    mock_config = {
+        "config_list": [{"model": "gpt-4o", "api_key": "fake", "base_url": "0.0.0.0:8000"}],
+        "temperature": 0,
+    }
     with patch("autogen.agentchat.conversable_agent.ConversableAgent.generate_oai_reply") as mock_oai_reply:
         agent = ReasoningAgent(
             "test_agent",

test/agentchat/test_agent_logging.py

@@ -50,7 +50,7 @@
     config_list = autogen.config_list_from_json(
         OAI_CONFIG_LIST,
         filter_dict={
-            "tags": ["gpt-3.5-turbo"],
+            "tags": ["gpt-4o-mini"],
         },
         file_location=KEY_LOC,
     )