Registry Overview

Through the Tiles Registry (opens in a new tab), users can download open weights models fine-tuned specifically for memory. These models use a human-readable external memory stored as markdown and learned policies trained via reinforcement learning on synthetically generated data to decide when to call Python functions that retrieve, update, or clarify that memory so the agent can maintain and refine persistent knowledge across sessions. We’re using Dria’s mem-agent research as inspiration, read their blog post, mem-agent: Equipping LLM Agents with Memory Using RL (opens in a new tab).

We are actively adding support for memory extensions with LoRA adapters, allowing users and organizations to bring their own data and augment the base memory models with the personality they want their memory to reflect.