LangChain v0.1.0 Launch: Retrieval

The best way to connect LLMs to your private data is currently retrieval augmented generation. LangChain has lots of advanced and production-ready functionality to help with this.
Jupyter Notebook (to follow along): github.com/hwchase17/langchai...
JavaScript Notebook: github.com/bracesproul/langch...
Links:
Retrieval Documentation: python.langchain.com/docs/mod...
Advanced Retrieval Methods: python.langchain.com/docs/mod...
QA with RAG Use Case Documentation: python.langchain.com/docs/use...

[@JDWilsonJr]

So appreciate the relentless refinement of the product and the documentation. It really helps.

[@colinmcnamara]

Really appreciate all the hard work that's been done improving the documentation. This is good stuff

[@leonardjin910]

LangChain is underrated for this 🙏

[@user-ze3sg6ix1u]

Thank you for this

[@arthursoenarto6051]

How does langchain’s retrieval compare with llamaindex’s?

[@RADKIT]

May i ask a question regarding doing a chain to ask a Json file?
i find it unclear how to do a retrieval > splitting (not sure if this is necessary in JSON) > embedding > vector store and then doing a chain to invoke a Q&A regarding a Json file, the documentation only covers to the data = loader.load() step
any help would be highly appreciated!

[@flaviobrienza6081]

Can I ask you why after the installation I cannot find langchain_openai?

[@LangChain]

its not installed by default, need to do `pip install langchain-openai`