[HELP]: AnythingLLM is not referring to the document uploaded to the workspace while answering questions

[BharathVibhav]

How are you running AnythingLLM?

Docker (local)

What happened?

Hi,
I'm running AnythingLLM with the following setup.

LanceDB

Macbook Pro 2018 6 Core i7
16GB RAM
512GB SSD
Docker

Running Ollama 3.2 on
Windows 11
AMD Ryzen 5 8 Core
16GB RAM
NVIDIA GeForce GTX 1650 Ti 2GB

I've scraped a website and have it's content in a CSV file (around 8MB). I have successfully uploaded the file to the workspace. but when I ask the chat for information about the contents in the file, the LLM is not getting it right.
Any help here would be appreciated.

Are there known steps to reproduce?

1. Upload a large CSV file
2. Use Ollama as the LLM
3. Ask the chat for information on the document's contents.
4. The chat comes up with no relevant information most of the times.

[timothycarambat]

We have a whole document on how the system works and how you can resolve these types of issues depending on your use case

https://docs.anythingllm.com/llm-not-using-my-docs