search

Mintplex-Labs / anything-llm

The all-in-one Desktop & Docker AI application with built-in RAG, AI agents, and more.
https://anythingllm.com
MIT License
24.04k stars 2.41k forks source link

[FEAT]: Limit Response to only the Custom Data provided #597

Closed bibutikoley closed 8 months ago

bibutikoley commented 8 months ago

What would you like to see?

I'm new to the world of AI and still exploring the features that this repo offers. Not sure if it works in the current version or not (If it does work, please let me know how can I do it)

Is it possible that LLM/Bot provides response only from the documents that is provided to it while embedding i.e. vector database. The bot should NOT use its own knowledge or data from internet. Please let me know how can this be achieved.

Thanks!

sumitsodhi88 commented 8 months ago

there are severe hallucinations from the model. also the model is not able to find and understand from the context. when asked about things it veer off into out of context. making the application unusable.

timothycarambat commented 8 months ago

@sumitsodhi88 This is a problem with your model or the settings of AnythingLLM workspace (such as threshold) and not something that this application explicitly controls.

sumitsodhi88 commented 8 months ago

@sumitsodhi88 This is a problem with your model or the settings of AnythingLLM workspace (such as threshold) and not something that this application explicitly controls.

yes, i shfted to (properlyformatted) word files and tinkered some tk get result. 👍