Persistent Memory for LLM - Personal AI - Artificial Brain

Braina works like an artificial brain. The aim of Braina (Brain Artificial) project is to develop a software with cognitive abilities similar to those of the human brain so that it can understand human language, think, infer, reason and learn. Braina does language understanding and learns from every conversation. It can also remember and recall the information which it has learned before. It evolves dynamically through long-term interactions with their users like a true personal assistant!

Not only Braina learns and remembers from conversations, it can also recall specific memory based on the context and it can pass it to large language models for further operation. Persistent memory for LLM allows you to use LLMs more robustly with great personalization! Personalized AI with long-term or perpetual memory, access to unlimited data and self-editing memory makes Braina a game-changer!

Important: Artificial Brain feature only works in Pro or higher version. You must be online and use Braina, Braina Swift or Braina Pinnacle Advanced AI Chat engines to access this feature. All data is stored on your own computer.

To use Braina's memory with LLM, you will need to enable "Persistent Memory context" feature. Click on the icon on the statusbar or enable persistent memory from Settings->Search/AI Engines tab. The icon will turn to blue suggesting that Persistent Memory is enabled.

Since Braina's memory can grow huge with time, you can specify a limit for amount of memory information that should be passed to the LLM to control usage of Advanced AI Credits. Go to Settings->Search/AI Engines tab and select the value under "Persistent Memory Input Context Limit (in words)". Default is 1500 words.

To remove some specific information from Braina's persistent memory system, simply delete your chat conversation(s) that contains that information and Braina will eventually forget it.