LangChain + MemorySync

Overview

Install langchain-memorysync (Python) or memorysync-langchain (Node.js) to give any LangChain chain or agent durable memory. The chat history class plugs into RunnableWithMessageHistory and keeps each session’s transcript in MemorySync conversation history, apart from the user’s memories; the durable facts in user messages are extracted into memories, and clear() deletes the transcript while those facts stay. A context provider returns prompt-ready recall blocks, and both packages run on langchain-core 0.3.x and 1.x.

Setup status and requirements

Supported version

langchain-memorysync 1.2.0 (PyPI) / memorysync-langchain 1.2.0 (npm)

Last setup review

2026-10-01

Permissions

A MemorySync API key with read and write scopes.

Limits

Requires langchain-core 0.3 or newer (Python 3.10+, Node 18+). History clear() needs a key with delete permission; anonymous evaluation keys can write and read history but cannot clear it.

Capabilities

• BaseChatMessageHistory implementation for RunnableWithMessageHistory
• Prompt-ready recall context blocks
• Transcript kept as written in conversation history — short answers survive, never listed as memories
• User messages sent to fact extraction, so durable facts become memories
• Idempotent writes that survive retries
• Per-end-user scoping and read-only mode
• langchain-core 0.3.x and 1.x support in both languages

Quick Start

from langchain_core.runnables.history import RunnableWithMessageHistory
from langchain_memorysync import MemorySyncChatMessageHistory

chain_with_history = RunnableWithMessageHistory(
    chain,
    lambda session_id: MemorySyncChatMessageHistory(session_id=session_id),
    input_messages_key="input",
    history_messages_key="history",
)

Use Cases

Explore More