supermemory is an open-source memory infrastructure for AI agents and applications. It provides persistent memory, semantic search, and knowledge graph capabilities that let agents remember context across sessions, retrieve relevant past interactions, and build up domain knowledge over time. Designed as a drop-in memory backend for coding agents, chatbots, and autonomous workflows.
supermemory provides the memory layer that most AI agents lack out of the box. While LLMs process each conversation in isolation, supermemory gives agents the ability to remember past interactions, store learned patterns, and retrieve relevant context from previous sessions. The project has grown to 25k+ GitHub stars with +501 stars/day velocity.
The engine combines vector storage for semantic similarity search with a knowledge graph for structured relationships. Agents can store observations, retrieve relevant memories by semantic query, and build up domain expertise over time. The API is framework-agnostic — it works with Claude Code, Codex, custom agents, and chatbot platforms.
supermemory deploys as a standalone service with a REST API. Self-hosted with Docker or available as a managed cloud service. Integrates with popular agent frameworks through simple API calls for store, query, and manage operations.
Builder.io's open-source framework for building agent-native applications — shared actions, SQL-backed state, identity, tools, skills, jobs, observability, and UI surfaces that all work together.
Persistent memory layer for AI coding agents — benchmark-backed (95.2% on LongMemEval-S), 92% fewer tokens per session vs full-context pasting, zero manual memory.add() calls.
Open-source AI pair programming tool that works in your terminal to edit code across your entire repository.