Preprint
Aug 2026
Total Recall at What Cost? Benchmarking the Serving Cost of Agentic Memory Systems
A break-even analysis shows that whether -- and when -- a memory system becomes cheaper to serve than the full transcript is highly sensitive to the system and the backbone, from the first tens of turns for the cheapest to never within 400 turns for the most expensive.
Natchanon Pollertlam, Witchayut Kornsuwannawit
· 1 citation