I would like to request the inclusion of a LangChain Node in n8n. LangChain is a library that assists in the development of language-based applications that combine large language models (LLMs) with other sources of computation or knowledge. The LangChain Node would enable n8n users to integrate language-based applications with their workflows.
My use case:
As a language-based application developer, I often use LLMs to build powerful applications. With the LangChain Node, I would be able to integrate LLMs with other sources of computation or knowledge, resulting in even more powerful and efficient applications. For example, I could use the LangChain Node to build a chatbot that uses an LLM to understand natural language and then integrates with other nodes in my workflow to perform tasks.
I think it would be beneficial to add this because:
The LangChain Node would add significant value to n8n by enabling users to build language-based applications that are more powerful and efficient than what is currently possible. With the increasing importance of natural language processing and understanding, the LangChain Node would be a valuable addition to n8n’s library of nodes.
Unfortunately, I am not skilled in node development and would not be able to contribute to the development of the LangChain Node. However, I am willing to provide feedback and testing support.
Thanks so much for the suggestion @Biscottino. We indeed already played around with a Langchain node in an internal hackathon to test out what it can do and how we can use it best. While we saw a huge opportunity for value, we also saw some risks - in paticular how early LangChain still is and that it can have unwanted destructive actions sometimes.
We will continue to explore. In the meantime, I would be super happy to hear more about potential use cases that you might have.
One workaround instead of depending/waiting on langchain to mature to be added is to add in the other sources/integrations that work with langchain:
Some exist already in n8n but most don’t. Like Vector DBs like Pinecone, Hugging Face and more…
Langchain is basically code version of n8n for ai. N8n has most* of the no-code/code integrations for tools/apps out there, it’s just the few version in there.
Samples usecases are creating chatbots, autogpt agents that run specifically based on your n8n nodes and customized data unlocking a lot of possibilities.
Thanks a lot, everybody! As an update. We are actively working on integrating LangChain capabilities into n8n already. We are looking forward to getting the first beta version released soon and then hopefully winning some of you as beta testers to receive some much-needed feedback!
I think what n8n really needs here, is the flowise/langflow integration. You dont need to create all those memory, template, index nodes from langchain because they are already implemented in 2 open source self hosted apps. Won’t worth it. What we really need here is a meaningful integration with those platforms where in one automation flow we can create a chat session with the langchain chat and be able to repeatedly call it
