Researchers introduced the Narrative World Model (NWM), a writer-memory system that uses a narratology-grounded typed temporal-state graph and query-conditioned hybrid retrieval to answer multi-hop qu…
Perseus Computing LLC released Mimir, a local-first, encrypted memory engine for AI agents as a single Rust binary with zero dependencies. The open-source tool provides persistent memory across sessio…
Zep released the Memory MCP Server, enabling any MCP client to access a unified user memory graph governed by enterprise single sign-on. The server allows desktop assistants, coding agents, and custom…
Microsoft Research unveiled Memora, a memory system for AI agents that decouples storage from retrieval to provide scalable long-term recall. The system reduces context token usage by up to 98% while …
Microsoft researchers introduced Memora, a scalable memory system for AI agents that decouples stored content from retrieval mechanisms, achieving state-of-the-art performance on long-horizon benchmar…
Truvem founder Dieng Amine has released TMX v0.1, an open standard JSON format for AI agent memory that enables portability across platforms, frameworks, and providers. The specification aims to solve…
Developers building production AI agents in 2026 are using simple vector search, keywords, BM25, text matching, and RRF for long-term memory, avoiding graph construction due to costs. One team reports…
Zep reports that placing agent memory in a trailing message instead of the system prompt can cut token costs by up to 2x by preserving prompt caching. The technique works on any provider and savings g…
Kage, an open-source framework for AI coding agents, introduces verified memory that is stored as files in a git repository, with citations checked against the codebase before writing and stale entrie…
Go's runtime is well-suited for production AI agents, which are long-running, concurrent, I/O-bound processes. The language's goroutines, channels, and single-binary deployment enable teams to build a…
Writer's research paper reports that memory systems can amplify sycophancy by up to 25x, but the amplification is traced to two design decisions: an evaluation prompt ordering the model to answer sole…
Zep launched the Batch API for loading large datasets into agent memory, supporting batches up to 50,000 items with a progress dashboard and no impact on real-time ingestion. The API replaces earlier …
A developer has found that AI agent memory systems optimized for retrieval accuracy can produce more unsafe actions than less accurate retrieval strategies, because relevance and authority are separat…
A developer building in public for 21 days discovered that every major agent memory API—including Mem0, Zep, Letta, and OpenAI's Assistants—is missing a lifecycle state field to mark facts as supersed…
A solo developer in Accra, Ghana, has launched AgentRAM, a memory API for AI agents that eliminates the need for vector databases or embedding pipelines. The service provides durable key-value memory …
Researchers have introduced "Behavioral Specification" as a new interpretive layer for AI memory, arguing that current systems optimized for factual recall fail to capture how a specific person proces…
The AI industry's approach to solving statelessness—by using centralized, server-side memory services—is fundamentally flawed for sensitive workflows. Instead, the article proposes that persistent con…
Zep announced Observations, a new context type that automatically discovers patterns and insights across Context Graphs, capturing behaviors, decisions, and preferences that emerge across multiple epi…