While streaming, AgentExecutor does not provide differentiation between LLM thoughts and final response

[vinay-netomi]

Checked other resources

• I added a very descriptive title to this question.
• I searched the LangChain documentation with the integrated search.
• I used the GitHub search to find a similar question and didn't find it.

Commit to Help

• I commit to help with one of those options 👆

Example Code

agent_executor = AgentExecutor( agent=agent, tools=tools, return_source_documents=True,
                return_intermediate_steps=True,
            )

async for event in executor.astream_events:
      print(event)

Description

I am trying to stream the response from Agent Executor using astream_events. But while sharing the events, there is no identifier between a LLM thought or Final response. I have shared both the events below. Is there any way to distinguish and prevent from streaming the LLM thoughts?

LLM Thought event:
{'event': 'on_chat_model_stream', 'name': 'ChatOpenAI', 'run_id': '822b6ca7-75fd-4279-9628-a9df383993ca', 'tags': ['seq:step:3'], 'metadata': {'user_id': '', 'tags': [']}, 'data': {'chunk': AIMessageChunk(content='I')}}

Final Response Event:

{'event': 'on_chat_model_stream',
 'name': 'ChatOpenAI',
 'run_id': '3a24c921-01de-4dab-9f36-45522f8a2416',
 'tags': ['seq:step:3'],
 'metadata': {'user_id': '',
  'tags': [']},
 'data': {'chunk': AIMessageChunk(content=' is')}}

System Info

langchain==0.1.9