Mem0
Long-term memory layer for AI agents and assistants
Mem0 is profiled here as a RAG Framework tool for engineering teams. Read about features, pricing, and how it compares to related options in the tools directory.
Description
Mem0 is an open-source memory layer for AI applications created in 2024 by Taranjeet Singh and Deshraj Yadav, who evolved it from their earlier Embedchain project. It extracts salient facts from conversations, consolidates them across vector, graph, and key-value stores, and retrieves the relevant ones at inference time, so agents remember users across sessions without stuffing full histories into context. The project's research paper reports accuracy gains and large token savings over full-context baselines on the LOCOMO long-conversation benchmark.
Key Capabilities:
Automatic fact extraction and memory consolidation from conversations
User, session, and agent-level memory scopes
Graph memory for relationships between entities
Semantic retrieval API for context injection at inference time
Integrations with LangGraph, CrewAI, and the Vercel AI SDK
Apache 2.0 license with a managed platform option
Alternative tools
- Docling
Open-source document conversion built for RAG pipelines
- Unstructured
Turn raw documents into LLM-ready structured data
- Voyage AI
Retrieval-optimized embedding and reranking models
- Chroma
Developer-first embedding database that runs anywhere
- Qdrant
Rust-based vector search engine with rich filtering
- Weaviate
Open-source vector database with native hybrid search