anh

> @dnanhkhoa could you please fix the server to switch from Query Parameter to POST payload? The current implementation cannot handle a long query parameter, hence this error happens. hi,...

Thank you for your clarification and making phonemes configurable is super helpful. I'll try your suggestion.

Suppose you are trying out https://github.com/hwchase17/chat-langchain * Run your langchain-server locally its api port will be 8000, its default ui is http://localhost:4173/ * Enable tracing=true in get_chain() in main.py *...

While it is being fix, you can use `ConversationalRetrievalChain` and monkey patch, line 10--18 from this: https://github.com/hwchase17/chat-langchain/blob/c1c89a8cd24ea91a844ab6f487accbe9e4f6e507/query_data.py

Did you switch to `ChatOpenAI` instead of `OpenAI`? ```diff - from langchain.llms import OpenAI + from langchain.chat_models import ChatOpenAI ``` then ```python question_gen_llm = ChatOpenAI( model_name="gpt-3.5-turbo", temperature=0, verbose=True, callback_manager=question_manager, )...