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
27.62k stars 2.79k forks source link

[FEAT]: The ability to route saved conversations to a scalable storage database #2583

Closed danielrosehill closed 3 weeks ago

danielrosehill commented 3 weeks ago

What would you like to see?

I'd love to be able to store conversations in a custom endpoint.

This might allow the conversation history to be mined and explored in parallel systems like wikis / knowledge management tools.

For this to work effectively, individual outputs would need to be batched into their constituent conversations.

Logical databases for storage might be Postgres or MongoDB.

timothycarambat commented 3 weeks ago

I'd love to be able to store conversations in a custom endpoint.

Could you not export the information from the API endpoint /v1/system/export-chats and upload to your db of choice?

As far as other databases go, this is capture in https://github.com/Mintplex-Labs/anything-llm/issues/2574

The reason for SQLite is it is zero config or set up and the docker setup flows down into the desktop app via the desktop branch it is critical we don't ask people installing a desktop app "Okay, now first set up a PostgreSQL db" - which is a nonstarter for that userbase.

There are some subtle differences between SQLite/PG, but they prisma schemas should be able to port for the most part