feat: streaming responses with function call

application/llm/google_ai.py

@@ -152,7 +152,15 @@ class GoogleLLM(BaseLLM):
             config=config,
         )
         for chunk in response:
-            if chunk.text is not None:
+            if hasattr(chunk, "candidates") and chunk.candidates:
+                for candidate in chunk.candidates:
+                    if candidate.content and candidate.content.parts:
+                        for part in candidate.content.parts:
+                            if part.function_call:
+                                yield part
+                            elif part.text:
+                                yield part.text
+            elif hasattr(chunk, "text"):
                 yield chunk.text
 
     def _supports_tools(self):

application/llm/openai.py

@@ -111,13 +111,24 @@ class OpenAILLM(BaseLLM):
         **kwargs,
     ):
         messages = self._clean_messages_openai(messages)
-        response = self.client.chat.completions.create(
-            model=model, messages=messages, stream=stream, **kwargs
-        )
+        if tools:
+            response = self.client.chat.completions.create(
+                model=model,
+                messages=messages,
+                stream=stream,
+                tools=tools,
+                **kwargs,
+            )
+        else:
+            response = self.client.chat.completions.create(
+                model=model, messages=messages, stream=stream, **kwargs
+            )
 
         for line in response:
             if line.choices[0].delta.content is not None:
                 yield line.choices[0].delta.content
+            else:
+                yield line.choices[0]
 
     def _supports_tools(self):
         return True