Memory Store: Cross-thread long-term memory system

1Review of memory types: short-term vs. long-term memory

In cognitive science, there are many types of memory. In AI applications, the most important distinction isShort-term memory (within-thread)andLong-term memory (cross-thread)。

2LangGraph Store core API: put/search/get

LangGraph Store is an open sourcePersistent key-value storage base class（BaseStore), providing three core methods to manage cross-thread memory.

Initialize InMemoryStore

import uuid
from langgraph.store.memory import InMemoryStore

# 创建内存中的 Store 实例（生产环境可替换为数据库后端）
in_memory_store = InMemoryStore()

put(): write to memory

# 定义命名空间：(user_id, 分类名)
user_id = "1"
namespace_for_memory = (user_id, "memories")

# 生成唯一 key（每条记忆一个 UUID）
key = str(uuid.uuid4())

# value 必须是字典
value = {"food_preference": "I like pizza"}

# 写入 Store
in_memory_store.put(namespace_for_memory, key, value)

search(): Batch search by namespace

# 检索命名空间下的所有记忆
memories = in_memory_store.search(namespace_for_memory)

# 每条记忆的完整结构示例：
# {
#   'value': {'food_preference': 'I like pizza'},
#   'key': 'a754b8c5-...',
#   'namespace': ['1', 'memories'],
#   'created_at': '2024-11-04T22:48:16.727572+00:00',
#   'updated_at': '2024-11-04T22:48:16.727574+00:00'
# }

get(): precise retrieval by namespace + key

# 精确获取单条记忆
memory = in_memory_store.get(namespace_for_memory, key)
memory.dict()  # 返回与 search()[0] 相同的结构

3Namespace design: organizing memory like a file system

Store'snamespaceIt is a tuple that works like a file system directory path, allowing you to hierarchically isolate different users and different types of memories.

# 命名空间：按 user_id 隔离，"memory" 是分类前缀
namespace = ("memory", user_id)

# 固定 key：每个用户只保存一条"汇总画像"记忆
key = "user_memory"

# 读取：先检查记忆是否存在
existing_memory = store.get(namespace, key)

if existing_memory:
    content = existing_memory.value.get('memory')
else:
    content = "No existing memory found."

4Chatbot architecture: responsibilities of two memory nodes

The whole picture is very simple:START → call_model → write_memory → END. Every time a user sends a message, it will go through two stages: "response" and "memory update".

5call_model node: reads memory from Store and personalizes response

MODEL_SYSTEM_MESSAGE = """You are a helpful assistant with memory that provides information about the user.
If you have memory for this user, use it to personalize your responses.
Here is the memory (it may be empty): {memory}"""

from langgraph.store.base import BaseStore
from langchain_core.runnables.config import RunnableConfig

def call_model(state: MessagesState, config: RunnableConfig, store: BaseStore):
    # Step 1：从 config 中取出 user_id
    user_id = config["configurable"]["user_id"]

    # Step 2：构造命名空间，从 Store 读取记忆
    namespace = ("memory", user_id)
    existing_memory = store.get(namespace, "user_memory")

    # Step 3：提取记忆内容（首次使用时为 None）
    if existing_memory:
        existing_memory_content = existing_memory.value.get('memory')
    else:
        existing_memory_content = "No existing memory found."

    # Step 4：将记忆注入系统提示词，调用 LLM
    system_msg = MODEL_SYSTEM_MESSAGE.format(memory=existing_memory_content)
    response = model.invoke(
        [SystemMessage(content=system_msg)] + state["messages"]
    )
    return {"messages": response}

6write_memory node: reflect on the conversation and update long-term memory

CREATE_MEMORY_INSTRUCTION = """You are collecting information about the user to personalize your responses.

CURRENT USER INFORMATION:
{memory}

INSTRUCTIONS:
1. Review the chat history below carefully
2. Identify new information about the user, such as:
   - Personal details (name, location)
   - Preferences (likes, dislikes)
   - Interests and hobbies
3. Merge any new information with existing memory
4. Format the memory as a clear, bulleted list
5. If new information conflicts with existing memory, keep the most recent version

Based on the chat history below, please update the user information:"""

def write_memory(state: MessagesState, config: RunnableConfig, store: BaseStore):
    user_id = config["configurable"]["user_id"]
    namespace = ("memory", user_id)
    existing_memory = store.get(namespace, "user_memory")

    if existing_memory:
        existing_memory_content = existing_memory.value.get('memory')
    else:
        existing_memory_content = "No existing memory found."

    # LLM 反思对话，提取并合并用户信息
    system_msg = CREATE_MEMORY_INSTRUCTION.format(memory=existing_memory_content)
    new_memory = model.invoke(
        [SystemMessage(content=system_msg)] + state['messages']
    )

    # 用相同 key 覆盖写入（合并更新）
    store.put(namespace, "user_memory", {"memory": new_memory.content})

7Dual storage compilation mode and configuration injection

Pictures need to be accepted at the same timecheckpointer(short term memory) andstore(Long Term Memory) compiled with two parameters.

from langgraph.checkpoint.memory import MemorySaver
from langgraph.graph import StateGraph, MessagesState, START, END
from langgraph.store.memory import InMemoryStore

builder = StateGraph(MessagesState)
builder.add_node("call_model", call_model)
builder.add_node("write_memory", write_memory)
builder.add_edge(START, "call_model")
builder.add_edge("call_model", "write_memory")
builder.add_edge("write_memory", END)

# 长期记忆：跨线程 Store
across_thread_memory = InMemoryStore()
# 短期记忆：线程内 Checkpointer
within_thread_memory = MemorySaver()

# 关键：同时传入 checkpointer（短期）和 store（长期）
graph = builder.compile(
    checkpointer=within_thread_memory,
    store=across_thread_memory
)

# thread_id → MemorySaver；user_id → Store
config = {
    "configurable": {
        "thread_id": "1",
        "user_id": "1"
    }
}

for chunk in graph.stream(
    {"messages": [HumanMessage(content="Hi, my name is Lance")]},
    config,
    stream_mode="values"
):
    chunk["messages"][-1].pretty_print()

8Full running demo: Cross-thread memory persistence

Thread 1: Establishing a user profile

Second conversation (Thread 2): Cross-thread memory takes effect

Open a newthread_id, but using the sameuser_id——The robot should be able to remember Lance’s information:

Design summary