Selective memory for AI agents importance-scored, self-decaying memory that forgets what doesn't matter so retrieval stays fast and relevant over long-running sessions.
The open-source benchmark for LLM memory decay. Measure how Naive, RAG, Chunked RAG, Cascading, and SummaryMemory degrade over 100 conversation turns. Ebbinghaus forgetting curves, 5-provider LLM eval, multi-seed CI. No API key needed.
