In the previous article, we learned about Retrieval-Augmented Generation (RAG). This has emerged as a powerful technique to enhance the capabilitie...
For further actions, you may consider blocking this person and/or reporting abuse
That is sooo useful! Saving it for future reference :D
Thanks Bap β€οΈ
Been messing with this stuff myself lately and man, getting it running local really feels solid. You think most people will go for local setups or just stick with the easy cloud stuff?
Cloud is easy and expensive. You pay for saving time and effort. Also, for completely private files and notes, better to setup locally. Should I create an app that works like that? For your local files?
What is the minimal configuration machine for running in local ?
Macbook M2 Pro Base model with 16 gb RAM.
Windows 11 + WSL with 16 GB RAM, i7 10th Gen, 2060.
Working fast in both the cases.
Nice! π₯
Thanks mate β€οΈ
Nice article!
when we are building local RAG, which is better to use for vector database, FAISS or pinecone?
What is pros and cons of using FIASS and pinecone?
For local RAG, and your data is around 2-3 GB (PDFs, Txt, Excels ,etc) and mostly static like your notes or something. Then you can easily use FAISS and it works pretty fine.
If the data crosses that, then getting good results will becomes a challenge. So, dedicated vector database might be helpful in that case. Also, Pinecone is pretty good, there are other like Qdrant, Weaviate, etc. which have a generous free tier.
Thanks for your reply.
been diving into rag setups myself and honestly, getting all this working local is such a flex. love how clean the ui is too, makes me want to tweak my own setup asap
Thanks Nathan. β€οΈ
learning a bit about rag, will definitely come back to this later. awesome tutorial Saurabh.
Thank you buddy!