Support for gpt-3.5-turbo-0125

Author: alexeichhorn

tests/test_openai.py

@@ -26,14 +26,15 @@ def test_token_counter(self):
         ]
 
         # check if test covers all models (increase if new models are added)
-        assert len(OpenAIChatModel.__args__) == 16  #  type: ignore
+        assert len(OpenAIChatModel.__args__) == 17  #  type: ignore
 
         client = AsyncTypeOpenAI(api_key="mock")
 
         assert client.chat.completions.num_tokens_from_messages(test_messages, model="gpt-3.5-turbo") == 27
         assert client.chat.completions.num_tokens_from_messages(test_messages, model="gpt-3.5-turbo-0301") == 29
         assert client.chat.completions.num_tokens_from_messages(test_messages, model="gpt-3.5-turbo-0613") == 27
         assert client.chat.completions.num_tokens_from_messages(test_messages, model="gpt-3.5-turbo-1106") == 27
+        assert client.chat.completions.num_tokens_from_messages(test_messages, model="gpt-3.5-turbo-0125") == 27
         assert client.chat.completions.num_tokens_from_messages(test_messages, model="gpt-3.5-turbo-16k") == 27
         assert client.chat.completions.num_tokens_from_messages(test_messages, model="gpt-3.5-turbo-16k-0613") == 27
         assert client.chat.completions.num_tokens_from_messages(test_messages, model="gpt-4") == 27
@@ -49,14 +50,15 @@ def test_token_counter(self):
 
     def test_max_token_counter(self):
         # check if test covers all models (increase if new models are added)
-        assert len(OpenAIChatModel.__args__) == 16  #  type: ignore
+        assert len(OpenAIChatModel.__args__) == 17  #  type: ignore
 
         client = AsyncTypeOpenAI(api_key="mock")
 
         assert client.chat.completions.max_tokens_of_model("gpt-3.5-turbo") == 4096
         assert client.chat.completions.max_tokens_of_model("gpt-3.5-turbo-0301") == 4096
         assert client.chat.completions.max_tokens_of_model("gpt-3.5-turbo-0613") == 4096
         assert client.chat.completions.max_tokens_of_model("gpt-3.5-turbo-1106") == 16384
+        assert client.chat.completions.max_tokens_of_model("gpt-3.5-turbo-0125") == 16384
         assert client.chat.completions.max_tokens_of_model("gpt-3.5-turbo-16k") == 16384
         assert client.chat.completions.max_tokens_of_model("gpt-3.5-turbo-16k-0613") == 16384
         assert client.chat.completions.max_tokens_of_model("gpt-4") == 8192

typegpt/openai/base_chat_completion.py

@@ -9,7 +9,7 @@ def max_tokens_of_model(model: OpenAIChatModel) -> int:
         match model:
             case "gpt-3.5-turbo" | "gpt-3.5-turbo-0301" | "gpt-3.5-turbo-0613":
                 return 4096
-            case "gpt-3.5-turbo-16k" | "gpt-3.5-turbo-16k-0613" | "gpt-3.5-turbo-1106":
+            case "gpt-3.5-turbo-16k" | "gpt-3.5-turbo-16k-0613" | "gpt-3.5-turbo-1106" | "gpt-3.5-turbo-0125":
                 return 16384
             case "gpt-4" | "gpt-4-0314" | "gpt-4-0613":
                 return 8192
@@ -46,6 +46,7 @@ def num_tokens_from_messages(cls, messages: list[EncodedMessage], model: OpenAIC
             "gpt-4-0613",
             "gpt-3.5-turbo-0613",
             "gpt-3.5-turbo-1106",
+            "gpt-3.5-turbo-0125",
             "gpt-3.5-turbo-16k-0613",
             "gpt-4-32k-0314",
             "gpt-4-32k-0613",

typegpt/openai/views.py

@@ -6,6 +6,7 @@
     "gpt-3.5-turbo-0301",
     "gpt-3.5-turbo-0613",
     "gpt-3.5-turbo-1106",
+    "gpt-3.5-turbo-0125",
     "gpt-3.5-turbo-16k",  # 3.5 turbo 16k
     "gpt-3.5-turbo-16k-0613",
     "gpt-4",  # gpt-4