Loading request...
A user has built a local AI assistant but is curious about adding features to improve memory retention during conversations. They want to know what features others would suggest for enhancing the memory approach in local AI systems.
I've been experimenting with running LLMs locally instead of relying on cloud APIs. Most setups require a lot of manual steps, so built a small installer that sets up a local AI environment automatically using: • llama.cpp • Dolphin 3.0 Llama 3.1 8B • persistent conversation memory • a simple launcher The goal was to make it easier to run a fully local AI assistant on your own machine. Everything runs locally and conversations never leave your system. Curious what people here think about the memory approach or what features you'd add to something like this.