langchain_openai[patch]: Invoke callback prior to yielding token (#18269

)

## PR title
langchain_openai[patch]: Invoke callback prior to yielding token

## PR message
Description: Invoke callback prior to yielding token in _stream and
_astream methods for langchain_openai.
Issue: #16913
Dependencies: None
Twitter handle: None

libs/partners/openai/langchain_openai/llms/base.py

@@ -253,7 +253,7 @@ def _stream(
             if not isinstance(stream_resp, dict):
                 stream_resp = stream_resp.model_dump()
             chunk = _stream_response_to_generation_chunk(stream_resp)
-            yield chunk
+
             if run_manager:
                 run_manager.on_llm_new_token(
                     chunk.text,
@@ -265,6 +265,7 @@ def _stream(
                         else None
                     ),
                 )
+            yield chunk
 
     async def _astream(
         self,
@@ -281,7 +282,7 @@ async def _astream(
             if not isinstance(stream_resp, dict):
                 stream_resp = stream_resp.model_dump()
             chunk = _stream_response_to_generation_chunk(stream_resp)
-            yield chunk
+
             if run_manager:
                 await run_manager.on_llm_new_token(
                     chunk.text,
@@ -293,6 +294,7 @@ async def _astream(
                         else None
                     ),
                 )
+            yield chunk
 
     def _generate(
         self,