L4 QA Chains and Calling the LLM via Langchain not working


Please if anyone can help with this, very frustrating:

Calling the LLM via Langchain is taking a lloooooong time, and when at the end, no results displayed as you can see in the attached picture.

I managed to get the desired result by calling the LLM directly, but the whole point of the course is to use Chains and Langchain


@Diala Diala can you please help me with this? could it be an issue of slow internet connection in Lebanon? Thank you

@Geoff @eddy can anyone help with this please?

1 Like

Can you please share your code and the error message?

Thanks for replying Diala, I already attached screenshots for the code, it is exactly the same code as the one in the lecture. There is no error, however calling the LLM via Langchain returns nothing and on the other hand, as you can see in the second picture, calling the LLM directly returns the desired result

1 Like

Experiencing same issue as @haydara009 . I also was able to get the desired result by importing openai and then calling LLM directly via openai.ChatCompletion.create and passing appropriate response element to display function. When running on my own local system notebook the lesson code “timed out” and no results were displayed. When running the code in the notebook embedded in the learn.deeplearning.ai page received 504 response error. Edit: I am in rural Maryland with decent residential broadband service.


Hi Eric, It’s sad that none of the instructors are addressing this issue and it’s in the heart of this course, the library is called langCHAIN and the whole point is to be able to use the CHAINs lol. I contacted community members and also have the same issue. If you found a solution for this kindly share with us @funkyspec

1 Like

Hello @haydara009 and @funkyspec, this is probably caused by the frequent updates in LangChain and an API change. I sent this to our engineering team. I will send you a follow up when this is fixed. Thank you for noticing this error.


Hello, is this issue fixed? I am facing the exactly same error.