In-browser LLM memory for React web apps. Bring your own SQLite adapter (e.g., sql.js WebAssembly) for a complete, zero-server RAG experience.

Alpha MCP server for local LLM memory, knowledge, state, routing, testing, and orchestration.

Local-first LLM memory for Expo and React Native. Combines the core semantic search and extraction engine with expo-sqlite and ready-to-use React hooks.

Platform-agnostic TypeScript engine for hybrid LLM memory. OKF v0.2 (llm-wiki/2) conformance with back-compat to v0.1 (llm-wiki/1) and profile-0 bundles. Episodic fact extraction, semantic vector search, multi-agent architectures over SQLite. Bring your o

Auditable local-LLM memory estimates for GPUs and Macs. Models sliding-window, hybrid/linear, MLA compressed-KV and MoE architectures. Zero deps, single file, conformance-vector tested.

Auditable local-LLM memory estimates for GPUs and Macs. Models sliding-window, hybrid/linear, MLA compressed-KV and MoE architectures. Zero deps, single file, conformance-vector tested.

Provider interfaces (LLM, Memory, Observability) for the EASA framework.

TypeScript/JavaScript client for LLM-Memory-Graph gRPC service

Neural LLM Memory MCP Server - High-performance neural memory system for AI assistants

Minimal, Pluggable LLM Memory SDK

Mastra is a framework for building AI-powered applications and agents with a modern TypeScript stack.

Universal LLM Memory Architecture (ULMA) plugin for OpenCode with project/session isolation

Local SQLite memory engine for Oh My Pi agents

LLM Memory

The Memory Layer For Your AI Apps

Self-hosted vector database + NoSQL with built-in AI semantic search — the Node.js & TypeScript client for Montycat. A Rust-powered, AI-native Pinecone / Weaviate / Chroma alternative for RAG, AI agents & LLM memory.

Universal LLM Memory Architecture (ULMA) plugin for OpenCode with project/session isolation

Node.js atomic and non-atomic counters, rate limiting tools, protection from DoS and brute-force attacks at scale

V5 Meta Memory System - Universal LLM Memory Framework

The write side of an LLM memory system: reconcile new claims against stored memory (confirm / supersede / contradict / independent) with cardinality, valid-time, calibrated confidence, and an audit trail.