Lucas
Lucas2mo ago

custom generation langchain runnable lambda

Hello, I'm using Lanchain (and Fireworks). I'm using the _generate method from ChatFireworks (to be able to set the nparameter) as follows:
RunnableLambda(lambda _messages: model._generate(_messages, **kwargs))
RunnableLambda(lambda _messages: model._generate(_messages, **kwargs))
This Runnable is a part of a chain which I called using invoke and set the langfuse callback. The problem is langfuse doesn't detect it's a generation. Anyone has a suggestion ?
3 Replies
Marc
Marc2mo ago
Can you share an example trace?
Lucas
Lucas2mo ago
No description
Marc
Marc2mo ago
thanks, can you create an issue on github for this? langfuse.com/issues