fix(llms): restore completion API for gpt-3.5-turbo-instruct (#610)

pandasai/llm/openai.py

@@ -29,7 +29,8 @@ class OpenAI(BaseOpenAI):
     The list of supported Chat models includes ["gpt-4", "gpt-4-0613", "gpt-4-32k",
      "gpt-4-32k-0613", "gpt-3.5-turbo", "gpt-3.5-turbo-16k", "gpt-3.5-turbo-0613",
      "gpt-3.5-turbo-16k-0613", "gpt-3.5-turbo-instruct"].
-
+    The list of supported Completion models includes "gpt-3.5-turbo-instruct" and
+     "text-davinci-003" (soon to be deprecated).
     """
 
     _supported_chat_models = [
@@ -41,8 +42,8 @@ class OpenAI(BaseOpenAI):
         "gpt-3.5-turbo-16k",
         "gpt-3.5-turbo-0613",
         "gpt-3.5-turbo-16k-0613",
-        "gpt-3.5-turbo-instruct",
     ]
+    _supported_completion_models = ["text-davinci-003", "gpt-3.5-turbo-instruct"]
 
     model: str = "gpt-3.5-turbo"
 
@@ -101,7 +102,9 @@ def call(self, instruction: AbstractPrompt, suffix: str = "") -> str:
         """
         self.last_prompt = instruction.to_string() + suffix
 
-        if self.model in self._supported_chat_models:
+        if self.model in self._supported_completion_models:
+            response = self.completion(self.last_prompt)
+        elif self.model in self._supported_chat_models:
             response = self.chat_completion(self.last_prompt)
         else:
             raise UnsupportedOpenAIModelError("Unsupported model")

tests/llms/test_openai.py

@@ -111,6 +111,13 @@ def test_call_with_unsupported_model(self, prompt):
             llm = OpenAI(api_token="test", model="not a model")
             llm.call(instruction=prompt)
 
+    def test_call_supported_completion_model(self, mocker, prompt):
+        openai = OpenAI(api_token="test", model="gpt-3.5-turbo-instruct")
+        mocker.patch.object(openai, "completion", return_value="response")
+
+        result = openai.call(instruction=prompt)
+        assert result == "response"
+
     def test_call_supported_chat_model(self, mocker, prompt):
         openai = OpenAI(api_token="test", model="gpt-4")
         mocker.patch.object(openai, "chat_completion", return_value="response")