memvid/memvid

Memory layer for AI Agents. Replace complex RAG pipelines with a serverless, single-file memory layer. Give your agents instant retrieval and long-term memory.

What it solves

Memvid provides a portable, serverless memory layer for AI agents, eliminating the need for complex RAG pipelines or external vector databases. It allows agents to maintain persistent, long-term memory that is stored in a single, portable file, making memory infrastructure-free and model-agnostic.

How it works

Memvid organizes AI memory as an append-only sequence of "Smart Frames"—immutable units containing content, timestamps, and metadata. These frames are packaged into a single .mv2 file that includes the data segments, full-text indices (Tantivy), vector indices (HNSW), and chronological ordering. This design enables fast retrieval, crash safety, and the ability to rewind or branch memory states without needing a database server.

Who it’s for

Developers building long-running AI agents, offline-first AI systems, enterprise knowledge bases, and auditable AI workflows who need a lightweight, persistent memory solution that can be carried across different environments.

Highlights

  • Single-File Storage: All data, embeddings, and search structures are contained in one .mv2 file with no sidecar files.
  • High Performance: Offers ultra-low latency (0.025ms P50) and high throughput for memory recall.
  • Multi-Modal Support: Includes features for PDF text extraction, CLIP visual embeddings for image search, and Whisper audio transcription.
  • Time-Travel Debugging: Supports querying past memory states and replaying or branching the memory timeline.
  • Local & Cloud Embeddings: Supports local ONNX models (like BGE-small) and cloud APIs (OpenAI).

Related

  • Project
    mem9-ai/mem9

    mem9 provides a persistent, searchable memory service for AI coding agents, offering hosted or self‑hosted deployments, multi‑agent sharing, hybrid vector/text search, smart ingest, and plugin integrations for major agent platforms.

  • Project
    MemTensor/memmy-agent

    Memmy is a local memory and agent runtime that provides persistent context across different AI coding assistants, allowing them to share project history and preferences.

  • Project
    TencentCloud/TencentDB-Agent-Memory

    TencentDB Agent Memory is an open‑source memory hub for LLM agents. It captures conversations, extracts reusable skills, builds a Wiki from docs, and indexes code into a CodeGraph. A central UI (Memory Hub) lets teams manage assets, set fine‑grained ACLs, and bind specific memory pieces to individual agents. Agents talk to a uniform proxy, so any framework (Claude Code, DeepSeek Harness, OpenClaw, etc.) can use the same memory without code changes. Benchmarks show a 59 % boost in persona retention. The repo provides one‑click Docker deployment, migration tools, and a roadmap for zero‑config cold starts and more integrations.

  • Project
    mnemosyne-oss/mnemosyne

    Mnemosyne is an open‑source, SQLite‑backed memory layer for LLM agents. It offers a local‑first, zero‑cloud solution with a Python SDK, a built‑in MCP server, and optional encrypted sync. The system stores facts, embeddings, and temporal triples using a three‑tier BEAM architecture (working, episodic, triple‑store) and performs hybrid vector‑keyword search inside SQLite. It integrates with many agent platforms (Claude Code, Cursor, OpenWebUI, Hermes, etc.) and runs with a single `pip install` and minimal resources.

  • Project
    MemTensor/MemOS

    MemOS 2.0 is a Memory Operating System for LLM‑based AI agents. It offers a unified API to store, retrieve, edit, and delete multi‑modal memories (text, images, tool traces) organized as a graph. The system can run as a hosted cloud service (API‑key access) or be self‑hosted via Docker (Neo4j + Qdrant) or a local SQLite plugin. Ready‑made plugins integrate MemOS with OpenClaw, Hermes, and DeepSeek Harness. Benchmarks show it outperforms existing commercial memory products. Licensed under Apache 2.0.