Hey guys! I was wondering if a RAG bot can be created using momen. I was hoping to create an AI assistant page for people asking questions, and only returning information that i provided (from pdfs and documents), kind of like a knowledge bank (without it being of the website itself).
Thanks!!
(and maybe there are some basic instructions or how to’s for this?) thanks again, im new to any sort of coding/tech.
Hi, welcome to Momen! 
Yes — you can build an RAG agent in Momen.
This means your AI can “look up” and use information stored in:
If you have PDFs or documents, you’ll first need to process them (for example, using a tool like LlamaIndex) and then store the processed data in your database or connect it through an API — Momen doesn’t read local files directly.
One of our users, Leon, built an RAG bot for his community, you can check out his video here: https://youtu.be/6L2FxJx7KZI?si=mnG4iALvuxrJTB1k
hey! i have one question, so in leons video, there is an option to input data (through uploading files, approx 50:43 time stamp in his video https://www.youtube.com/watch?v=6L2FxJx7KZI). For me, i do not have that option in the context subsection. Is it apart of his Momen subscription plan? or was there something i need to download? (i am currently on free plan)
Thank you so much for your response!!
Hi, we no longer support uploading files directly into Momen (including the upload option you saw in that video). If you want to use your own PDFs, documents, or other data, you’ll first need to process them yourself — for example, using a tool like LlamaIndex — and then store the processed data in your own database or connect it to Momen via an API. This way, Momen can access the data without requiring file uploads.
The reason was that it was a poorly thought-out feature.
Ohh, i see
so currently, im using LlamaCloud to create an index, after that i would need to grab a llamacloud api and put in the Llamacloud api into Momen “API” in the context section, correct?
also what chunk size/chunk overlap do i select for LlamaCloud index? (Sorry i know its a momen forum)
If possible is there a video that demonstrates how i can do this? I would greatly appreciate it!
p.s. i really really appreciate all your guy’s responses (im new to all this 
)
Yes, you put llama cloud’s api into momen AI’s context.
You can also export the chunks and import them each as a row into a momen database table and turn on vector storage for the column containing the chunk’s content. And do the vector search on Momen’s side. It is cheaper but less flexible than llamacloud.
Thanks for your response,
i have having some trouble creating a momen table,
when trying to upload (import) my chunks to the momen data base, i cant import it to anything. Import to: (please select) and then in the dropdown menu there is “no data.”
i have tried creating a data model, however nothing is showing up, and im not able to input my data chunks
Ohhh i see i will try that! thank you
That worked thank you so much!!!
Why don’t have this in Momen docs? or do we and I am missing it? It would be super helpful to go to Getting Started with Momen - Complete Development Framework | Momen Docs and ask how to do something and it relieves support load and it’s a quicker response than human customer support.
We should, totally.
Never gotten around to it yet.
