Get openai request ID from langchain

[sanjeevanahilan]

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

llm: ChatOpenAI
output = await llm.ainvoke

Description

When calling ainvoke or agenerate I don't receive in the output the OpenAI request id. e.g. chatcmpl-9IaVqJEDF1IxZgiVui6YZmqNMEktY

This would come from the full response object. I need to track the id of each request for my application - how can I do this? Is there an optional argument I can provide?

System Info

langchain==0.0.343

[eyurtsev]

Upgrade to a new version of langchain and use the openai partner package. 0.0.343 is very old now

[rainmanwy]

I could confirm that when stream=True, the id is set by langchain(with "run-" prefix). and stream=False, id is set with id of openai api correctly.

I am using the latest langchain & langchain-openai version
langchain == 0.3.25
langchain-openai == 0.3.18

in "langchain_openai/chat_models/base.py"

def _convert_delta_to_message_chunk(
    _dict: Mapping[str, Any], default_class: type[BaseMessageChunk]
) -> BaseMessageChunk:
    id_ = _dict.get("id")
    role = cast(str, _dict.get("role"))
    content = cast(str, _dict.get("content") or "")
    additional_kwargs: dict = {}

with stream mode, langchain-openai try to get id in chunk, but actually "id" should be get from response as i think.

[thewebscraping]

you are right run-uuid4