Evaluation ReportActive
The LoCoMo Benchmark Report
Our core evaluation matrix checking long-context recall, entity traversal, and temporal consistency across 1,540 simulated developer query runs.
ACCURACY: 73.5% | HIT@10: 88.8% Read Report
/ RESEARCH & STUDIES
The Research Hub
Advancing AI Long-Term Memory.
Memwyre Research Lab publishes studies, evaluations, and datasets exploring how entity graphs, pruning algorithms, and context compression solve statelessness in AI agent networks.
Advance Your AI Architecture
Access our open source benchmark repositories, integrate Model Context Protocol memory servers, and build stateful agents.