SAGE
MemoryPersistent, consensus-validated memory infrastructure for AI agents, with confidence scoring and natural memory decay.
A FOCUSED STARTING POINT
Memory
Explore ways to carry useful context between agent sessions. Decide what should be remembered, how it can be corrected, and when it should be forgotten before choosing a storage layer.
What to compare
- Is the information a user preference, a fact or a temporary task state?
- Can a user inspect, correct and delete a memory?
- Will memories remain isolated between users?
Screenpipe
AI agentsScreenpipe captures screen and audio locally, then exposes searchable work history and agent context through an MCP server. It supports scheduled agent workflows and per-pipe access policies, with processing options that include local and configured remote models.
somora
AI agentsA self-hosted server for personal AI agents with persistent private memory, shared wiki, delegated agents, scheduled work, and web, shell, and browser tools across terminal, browser, and mobile clients.
Sonzai Go SDK
MemoryAn open-source Go client for the hosted Sonzai Mind Layer API, providing agent runtimes with memory context, turn persistence, and voice services.
TeamHero
Agent frameworksA self-hosted workspace for coordinating AI agent teams with structured tasks, persistent memory, and file-conflict controls.
Tree Ring Memory is a framework-agnostic, local-first memory lifecycle for AI agents, with SQLite search, deliberate forgetting, audit logs, consolidation, and runtime adapters.
Zep
MemoryZep is a long-term memory service for AI assistants that enhances recall, understanding, and data extraction from chat histories to power personalized AI experiences
Before you choose
Agent memory vs RAG: decide what should be remembered
Separate task state, user preferences and source documents, then choose storage, retrieval and deletion rules for each type of information.
Read the guide