MemOSMemTensor's memory OS for LLM agents: one API over graph-structured, multi-modal memory with hybrid retrieval and skill evolution — hosted, self-hosted (Neo4j + Qdrant) or local plugins.
Why switchBoth are full memory stacks with user-level recall, hybrid retrieval and agent plugins; MemOS emphasizes memory cubes and skill evolution, Supermemory fact extraction, user profiles and connectors.
Full comparison → hindsightAgent memory that learns, not just recalls: retain/recall/reflect API over Postgres, SOTA on LongMemEval. Self-host via Docker with UI; Python/TS clients, any LLM provider.
Why switchBoth are memory APIs chasing LongMemEval and LoCoMo; hindsight is a lean self-hosted retain/recall/reflect service on Postgres, Supermemory a broader platform with connectors, profiles and a one-binary local mode.
Full comparison → MemMachineLong-term memory layer for AI agents — episodic (graph), profile (SQL) and working memory behind Python/TS SDKs, REST and MCP; ships LangChain, LangGraph, CrewAI and LlamaIndex integrations.
Why switchSame job, long-term memory behind SDKs, REST and MCP: memmachine separates episodic, profile and working memory with framework integrations, Supermemory fuses memory with RAG and connectors.
Full comparison → claude-memCross-harness session memory: hooks capture what the agent does, an LLM compresses it into observations, and the next session gets the relevant ones back via progressive-disclosure MCP tools.
Why switchFor Claude Code specifically: claude-mem compresses the agent's own sessions into observations, Supermemory's plugin keeps preferences and project facts in a memory API shared across tools.
Full comparison →