Tiered memory managed by the model
Keep a small working context and let the model page facts in and out of external storage.
Treat the context window as main memory and an external store as disk. The model uses tools to write durable facts out, search them, and pull them back in when needed. The working context stays small and relevant across sessions of any length.
Contexts: Chat assistant, Autonomous agent, Coding agent
Does it work?
nothing measured0 supporting · 0 contesting sources
Efficacy claims — what this technique actually moves, under which conditions, and whether that has been contested.
1 more condition from claims reviewed by AI, listed with the claims below.
No efficacy claim filed yet. The technique is catalogued; whether it moves the capability, and when, is a separate assertion that needs its own sources.
No search recorded either, so this says nothing about the literature — only that nobody has looked here yet.
Code
- https://github.com/letta-ai/letta — Letta, formerly MemGPT, the reference implementation.verified 2026-09-02
Sources
Whether a technique works is a claim. What you file here becomes a claim about this technique under the first capability it addresses, with your evidence as its source, and moves the standing above the same way a paper would.