You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Is your feature request related to a problem? Please describe.
I have done qutie a bit of work on MemGPT in the past but there was never a satisfactory result with getting a lot background info into an agent before it started being asked questions - I tried all the suggested methods but, at best, an agent only "remembered" the information in a painfully slow way and only after lots of prompting about possible answers . . It is not viable to spend ages "training" an agent in this way - it should know everything, straight away that is in the pre-prepared data text file for the agent to be able to sensibly answer questions from members of my non-profit organisation.
Describe the solution you'd like
That things have improved a lot and I can now do what I need to be able to do - my last efforts were quite a while before Letta became a "thing" so I am hoping that is the case . .
The text was updated successfully, but these errors were encountered:
Is your feature request related to a problem? Please describe.
I have done qutie a bit of work on MemGPT in the past but there was never a satisfactory result with getting a lot background info into an agent before it started being asked questions - I tried all the suggested methods but, at best, an agent only "remembered" the information in a painfully slow way and only after lots of prompting about possible answers . . It is not viable to spend ages "training" an agent in this way - it should know everything, straight away that is in the pre-prepared data text file for the agent to be able to sensibly answer questions from members of my non-profit organisation.
Describe the solution you'd like
That things have improved a lot and I can now do what I need to be able to do - my last efforts were quite a while before Letta became a "thing" so I am hoping that is the case . .
The text was updated successfully, but these errors were encountered: