community[patch]: invoke callback prior to yielding token (openai) (l…

…angchain-ai#19389)

**Description:** Invoke callback prior to yielding token for BaseOpenAI
& OpenAIChat
**Issue:** [Callback for on_llm_new_token should be invoked before the
token is yielded by the model
langchain-ai#16913](langchain-ai#16913)
**Dependencies:** None

libs/community/langchain_community/llms/openai.py

@@ -391,7 +391,6 @@ async def _astream(
             if not isinstance(stream_resp, dict):
                 stream_resp = stream_resp.dict()
             chunk = _stream_response_to_generation_chunk(stream_resp)
-            yield chunk
             if run_manager:
                 await run_manager.on_llm_new_token(
                     chunk.text,
@@ -401,6 +400,7 @@ async def _astream(
                     if chunk.generation_info
                     else None,
                 )
+            yield chunk
 
     def _generate(
         self,
@@ -1113,9 +1113,9 @@ def _stream(
                 stream_resp = stream_resp.dict()
             token = stream_resp["choices"][0]["delta"].get("content", "")
             chunk = GenerationChunk(text=token)
-            yield chunk
             if run_manager:
                 run_manager.on_llm_new_token(token, chunk=chunk)
+            yield chunk
 
     async def _astream(
         self,
@@ -1133,9 +1133,9 @@ async def _astream(
                 stream_resp = stream_resp.dict()
             token = stream_resp["choices"][0]["delta"].get("content", "")
             chunk = GenerationChunk(text=token)
-            yield chunk
             if run_manager:
                 await run_manager.on_llm_new_token(token, chunk=chunk)
+            yield chunk
 
     def _generate(
         self,