Oracle AI Agent Memory 26.6 is built for high-accuracy, low-latency, enterprise-grade memory for agents. It introduces hybrid search, combining vector search for semantic understanding, keyword search for exact language, and scoped retrieval across users, agents, threads, record types, and metadata. It also supports durable memories, facts, guidelines, and preferences as distinct record types. Oracle AI Agent Memory runs on Oracle AI Database, where memory, metadata, scopes, transactional state, vector search, keyword search, hybrid indexes, retention controls, and deletion workflows can live together. With OracleDBEmbedder, embeddings can be generated in the database, reducing external network hops. For latency-sensitive applications, Agent Memory 26.6 supports vector-only retrieval, keyword retrieval, hybrid search, configurable index synchronization, and background memory extraction. The system also supports full CRUD, ensuring that "delete" should mean delete.