AI 应用需要记忆来在多次交互中共享上下文。在 LangGraph 中，你可以添加两种类型的记忆：

​
添加短期记忆

短期记忆（线程级持久化）让智能体能够追踪多轮对话。要添加短期记忆： 
from langgraph.checkpoint.memory import InMemorySaver  
from langgraph.graph import StateGraph

checkpointer = InMemorySaver()

builder = StateGraph(...)
graph = builder.compile(checkpointer=checkpointer)

graph.invoke(
    {"messages": [{"role": "user", "content": "hi! i am Bob"}]},
    {"configurable": {"thread_id": "1"}},
)

​
在生产环境中使用

在生产环境中，应使用由数据库支持的检查点存储器： 
from langgraph.checkpoint.postgres import PostgresSaver

DB_URI = "postgresql://postgres:postgres@localhost:5432/postgres?sslmode=disable"
with PostgresSaver.from_conn_string(DB_URI) as checkpointer:
    builder = StateGraph(...)
    graph = builder.compile(checkpointer=checkpointer)

pip install -U "psycopg[binary,pool]" langgraph langgraph-checkpoint-postgres
你在首次使用 Postgres 检查点存储器时需要调用 checkpointer.setup()。
from langchain.chat_models import init_chat_model
from langgraph.graph import StateGraph, MessagesState, START
from langgraph.checkpoint.postgres import PostgresSaver  

model = init_chat_model(model="claude-haiku-4-5-20251001")

DB_URI = "postgresql://postgres:postgres@localhost:5432/postgres?sslmode=disable"
with PostgresSaver.from_conn_string(DB_URI) as checkpointer:
    # checkpointer.setup()

    def call_model(state: MessagesState):
        response = model.invoke(state["messages"])
        return {"messages": response}

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

    graph = builder.compile(checkpointer=checkpointer)

    config = {
        "configurable": {
            "thread_id": "1"
        }
    }

    for chunk in graph.stream(
        {"messages": [{"role": "user", "content": "hi! I'm bob"}]},
        config,
        stream_mode="values"
    ):
        chunk["messages"][-1].pretty_print()

    for chunk in graph.stream(
        {"messages": [{"role": "user", "content": "what's my name?"}]},
        config,
        stream_mode="values"
    ):
        chunk["messages"][-1].pretty_print()

pip install -U pymongo langgraph langgraph-checkpoint-mongodb
环境搭建 使用 MongoDB 检查点存储器 你需要一个 MongoDB 集群。如果还没有，请按照本指南创建一个集群。
from langchain.chat_models import init_chat_model
from langgraph.graph import StateGraph, MessagesState, START
from langgraph.checkpoint.mongodb import MongoDBSaver  

model = init_chat_model(model="claude-haiku-4-5-20251001")

MONGODB_URI = "localhost:27017"
with MongoDBSaver.from_conn_string(MONGODB_URI) as checkpointer:

    def call_model(state: MessagesState):
        response = model.invoke(state["messages"])
        return {"messages": response}

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

    graph = builder.compile(checkpointer=checkpointer)

    config = {
        "configurable": {
            "thread_id": "1"
        }
    }

    for chunk in graph.stream(
        {"messages": [{"role": "user", "content": "hi! I'm bob"}]},
        config,
        stream_mode="values"
    ):
        chunk["messages"][-1].pretty_print()

    for chunk in graph.stream(
        {"messages": [{"role": "user", "content": "what's my name?"}]},
        config,
        stream_mode="values"
    ):
        chunk["messages"][-1].pretty_print()

pip install -U langgraph langgraph-checkpoint-redis
你在首次使用 Redis 检查点存储器时需要调用 checkpointer.setup()。
from langchain.chat_models import init_chat_model
from langgraph.graph import StateGraph, MessagesState, START
from langgraph.checkpoint.redis import RedisSaver  

model = init_chat_model(model="claude-haiku-4-5-20251001")

DB_URI = "redis://localhost:6379"
with RedisSaver.from_conn_string(DB_URI) as checkpointer:
    # checkpointer.setup()

    def call_model(state: MessagesState):
        response = model.invoke(state["messages"])
        return {"messages": response}

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

    graph = builder.compile(checkpointer=checkpointer)

    config = {
        "configurable": {
            "thread_id": "1"
        }
    }

    for chunk in graph.stream(
        {"messages": [{"role": "user", "content": "hi! I'm bob"}]},
        config,
        stream_mode="values"
    ):
        chunk["messages"][-1].pretty_print()

    for chunk in graph.stream(
        {"messages": [{"role": "user", "content": "what's my name?"}]},
        config,
        stream_mode="values"
    ):
        chunk["messages"][-1].pretty_print()

pip install -U langgraph langgraph-oracledb
环境搭建 使用 Oracle 检查点存储器，你需要一个 Oracle AI 数据库实例。本地容器（例如 gvenzl/oracle-free:23-slim）或 OCI 中的 Oracle 自治数据库都可以使用。
你在首次使用 Oracle 检查点存储器时需要调用 checkpointer.setup()。
from langchain.chat_models import init_chat_model
from langgraph.graph import StateGraph, MessagesState, START
from langgraph_oracledb.checkpoint.oracle import OracleSaver  

model = init_chat_model(model="claude-haiku-4-5-20251001")

DB_URI = "user/password@localhost:1521/FREEPDB1"
with OracleSaver.from_conn_string(DB_URI) as checkpointer:
    # checkpointer.setup()

    def call_model(state: MessagesState):
        response = model.invoke(state["messages"])
        return {"messages": response}

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

    graph = builder.compile(checkpointer=checkpointer)

    config = {
        "configurable": {
            "thread_id": "1"
        }
    }

    for chunk in graph.stream(
        {"messages": [{"role": "user", "content": "hi! I'm bob"}]},
        config,
        stream_mode="values"
    ):
        chunk["messages"][-1].pretty_print()

    for chunk in graph.stream(
        {"messages": [{"role": "user", "content": "what's my name?"}]},
        config,
        stream_mode="values"
    ):
        chunk["messages"][-1].pretty_print()

​
在子图中使用

如果你的图包含子图，你只需要在编译父图时提供检查点存储器。LangGraph 会自动将检查点存储器传播到子图。 
from langgraph.graph import START, StateGraph
from langgraph.checkpoint.memory import InMemorySaver
from typing import TypedDict

class State(TypedDict):
    foo: str

# 子图

def subgraph_node_1(state: State):
    return {"foo": state["foo"] + "bar"}

subgraph_builder = StateGraph(State)
subgraph_builder.add_node(subgraph_node_1)
subgraph_builder.add_edge(START, "subgraph_node_1")
subgraph = subgraph_builder.compile()

# 父图

builder = StateGraph(State)
builder.add_node("node_1", subgraph)
builder.add_edge(START, "node_1")

checkpointer = InMemorySaver()
graph = builder.compile(checkpointer=checkpointer)
你可以配置特定于子图的检查点行为。有关持久化级别（包括中断支持和有状态续接）的详细信息，请参阅子图持久化。 
subgraph_builder = StateGraph(...)
subgraph = subgraph_builder.compile(checkpointer=True)

​
添加长期记忆

使用长期记忆在对话之间存储用户特定或应用特定的数据。 
from langgraph.store.memory import InMemoryStore  
from langgraph.graph import StateGraph

store = InMemoryStore()

builder = StateGraph(...)
graph = builder.compile(store=store)

​
在节点内部访问存储器

一旦编译了带有存储器的图，LangGraph 会自动将存储器注入到你的节点函数中。推荐通过 Runtime 对象访问存储器。 
from dataclasses import dataclass
from langgraph.runtime import Runtime
from langgraph.graph import StateGraph, MessagesState, START
import uuid

@dataclass
class Context:
    user_id: str

async def call_model(state: MessagesState, runtime: Runtime[Context]):
    user_id = runtime.context.user_id  
    namespace = (user_id, "memories")

    # 搜索相关记忆
    memories = await runtime.store.asearch(
        namespace, query=state["messages"][-1].content, limit=3
    )
    info = "\n".join([d.value["data"] for d in memories])

    # ... 在模型调用中使用记忆

    # 存储一条新记忆
    await runtime.store.aput(
        namespace, str(uuid.uuid4()), {"data": "User prefers dark mode"}
    )

builder = StateGraph(MessagesState, context_schema=Context)
builder.add_node(call_model)
builder.add_edge(START, "call_model")
graph = builder.compile(store=store)

# 在调用时传递上下文
graph.invoke(
    {"messages": [{"role": "user", "content": "hi"}]},
    {"configurable": {"thread_id": "1"}},
    context=Context(user_id="1"),
)

​
在生产环境中使用

在生产环境中，应使用由数据库支持的存储器： 
from langgraph.store.postgres import PostgresStore

DB_URI = "postgresql://postgres:postgres@localhost:5432/postgres?sslmode=disable"
with PostgresStore.from_conn_string(DB_URI) as store:
    builder = StateGraph(...)
    graph = builder.compile(store=store)

pip install -U "psycopg[binary,pool]" langgraph langgraph-checkpoint-postgres
你在首次使用 Postgres 存储时需要调用 store.setup()。
from dataclasses import dataclass
from langchain.chat_models import init_chat_model
from langgraph.graph import StateGraph, MessagesState, START
from langgraph.checkpoint.postgres.aio import AsyncPostgresSaver
from langgraph.store.postgres.aio import AsyncPostgresStore  
from langgraph.runtime import Runtime  
import uuid

model = init_chat_model(model="claude-haiku-4-5-20251001")

@dataclass
class Context:
    user_id: str

async def call_model(
    state: MessagesState,
    runtime: Runtime[Context],
):
    user_id = runtime.context.user_id  
    namespace = ("memories", user_id)
    memories = await runtime.store.asearch(namespace, query=str(state["messages"][-1].content))
    info = "\n".join([d.value["data"] for d in memories])
    system_msg = f"You are a helpful assistant talking to the user. User info: {info}"

    # 如果用户要求模型记住某事，则存储新记忆
    last_message = state["messages"][-1]
    if "remember" in last_message.content.lower():
        memory = "User name is Bob"
        await runtime.store.aput(namespace, str(uuid.uuid4()), {"data": memory})

    response = await model.ainvoke(
        [{"role": "system", "content": system_msg}] + state["messages"]
    )
    return {"messages": response}

DB_URI = "postgresql://postgres:postgres@localhost:5432/postgres?sslmode=disable"

async with (
    AsyncPostgresStore.from_conn_string(DB_URI) as store,
    AsyncPostgresSaver.from_conn_string(DB_URI) as checkpointer,
):
    # await store.setup()
    # await checkpointer.setup()

    builder = StateGraph(MessagesState, context_schema=Context)
    builder.add_node(call_model)
    builder.add_edge(START, "call_model")

    graph = builder.compile(
        checkpointer=checkpointer,
        store=store,
    )

    config = {"configurable": {"thread_id": "1"}}
    async for chunk in graph.astream(
        {"messages": [{"role": "user", "content": "Hi! Remember: my name is Bob"}]},
        config,
        stream_mode="values",
        context=Context(user_id="1"),
    ):
        chunk["messages"][-1].pretty_print()

    config = {"configurable": {"thread_id": "2"}}
    async for chunk in graph.astream(
        {"messages": [{"role": "user", "content": "what is my name?"}]},
        config,
        stream_mode="values",
        context=Context(user_id="1"),
    ):
        chunk["messages"][-1].pretty_print()

pip install -U langgraph langgraph-checkpoint-redis
你在首次使用 Redis 存储 时需要调用 store.setup()。
from dataclasses import dataclass
from langchain.chat_models import init_chat_model
from langgraph.graph import StateGraph, MessagesState, START
from langgraph.checkpoint.redis.aio import AsyncRedisSaver
from langgraph.store.redis.aio import AsyncRedisStore  
from langgraph.runtime import Runtime  
import uuid

model = init_chat_model(model="claude-haiku-4-5-20251001")

@dataclass
class Context:
    user_id: str

async def call_model(
    state: MessagesState,
    runtime: Runtime[Context],
):
    user_id = runtime.context.user_id  
    namespace = ("memories", user_id)
    memories = await runtime.store.asearch(namespace, query=str(state["messages"][-1].content))
    info = "\n".join([d.value["data"] for d in memories])
    system_msg = f"You are a helpful assistant talking to the user. User info: {info}"

    # 如果用户要求模型记住某事，则存储新记忆
    last_message = state["messages"][-1]
    if "remember" in last_message.content.lower():
        memory = "User name is Bob"
        await runtime.store.aput(namespace, str(uuid.uuid4()), {"data": memory})

    response = await model.ainvoke(
        [{"role": "system", "content": system_msg}] + state["messages"]
    )
    return {"messages": response}

DB_URI = "redis://localhost:6379"

async with (
    AsyncRedisStore.from_conn_string(DB_URI) as store,
    AsyncRedisSaver.from_conn_string(DB_URI) as checkpointer,
):
    # await store.setup()
    # await checkpointer.asetup()

    builder = StateGraph(MessagesState, context_schema=Context)
    builder.add_node(call_model)
    builder.add_edge(START, "call_model")

    graph = builder.compile(
        checkpointer=checkpointer,
        store=store,
    )

    config = {"configurable": {"thread_id": "1"}}
    async for chunk in graph.astream(
        {"messages": [{"role": "user", "content": "Hi! Remember: my name is Bob"}]},
        config,
        stream_mode="values",
        context=Context(user_id="1"),
    ):
        chunk["messages"][-1].pretty_print()

    config = {"configurable": {"thread_id": "2"}}
    async for chunk in graph.astream(
        {"messages": [{"role": "user", "content": "what is my name?"}]},
        config,
        stream_mode="values",
        context=Context(user_id="1"),
    ):
        chunk["messages"][-1].pretty_print()

pip install -U langgraph langgraph-oracledb langchain-openai
环境搭建 使用 Oracle 存储，你需要一个 Oracle AI 数据库实例——语义 search 所需的向量索引需要 Oracle AI 向量搜索。
你在首次使用 Oracle 存储和检查点存储器时需要调用 store.setup() 和 checkpointer.setup()。
import uuid

from langchain.chat_models import init_chat_model
from langchain.embeddings import init_embeddings
from langchain_core.runnables import RunnableConfig
from langgraph.graph import StateGraph, MessagesState, START
from langgraph.store.base import BaseStore
from langgraph_oracledb.checkpoint.oracle import OracleSaver
from langgraph_oracledb.store.oracle import OracleStore  

model = init_chat_model(model="claude-haiku-4-5-20251001")
embeddings = init_embeddings("openai:text-embedding-3-small")

DB_URI = "user/password@localhost:1521/FREEPDB1"

with (
    OracleStore.from_conn_string(
        DB_URI,
        index={"embed": embeddings, "dims": 1536},
    ) as store,
    OracleSaver.from_conn_string(DB_URI) as checkpointer,
):
    store.setup()
    checkpointer.setup()

    def call_model(
        state: MessagesState,
        config: RunnableConfig,
        *,
        store: BaseStore,
    ):
        user_id = config["configurable"]["user_id"]
        namespace = ("memories", user_id)
        memories = store.search(namespace, query=str(state["messages"][-1].content))
        info = "\n".join([d.value["data"] for d in memories])
        system_msg = f"You are a helpful assistant talking to the user. User info: {info}"

        # 如果用户要求模型记住某事，则存储新记忆
        last_message = state["messages"][-1]
        if "remember" in last_message.content.lower():
            memory = "User name is Bob"
            store.put(namespace, str(uuid.uuid4()), {"data": memory})

        response = model.invoke(
            [{"role": "system", "content": system_msg}] + state["messages"]
        )
        return {"messages": response}

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

    graph = builder.compile(
        checkpointer=checkpointer,
        store=store,
    )

    config = {
        "configurable": {
            "thread_id": "1",
            "user_id": "1",
        }
    }
    for chunk in graph.stream(
        {"messages": [{"role": "user", "content": "Hi! Remember: my name is Bob"}]},
        config,
        stream_mode="values",
    ):
        chunk["messages"][-1].pretty_print()

    config = {
        "configurable": {
            "thread_id": "2",
            "user_id": "1",
        }
    }

    for chunk in graph.stream(
        {"messages": [{"role": "user", "content": "what is my name?"}]},
        config,
        stream_mode="values",
    ):
        chunk["messages"][-1].pretty_print()

​
使用语义搜索

在图的内存存储中启用语义搜索，使图智能体能够通过语义相似性在存储中搜索项目。 
from langchain.embeddings import init_embeddings
from langgraph.store.memory import InMemoryStore

# 创建启用语义搜索的存储
embeddings = init_embeddings("openai:text-embedding-3-small")
store = InMemoryStore(
    index={
        "embed": embeddings,
        "dims": 1536,
    }
)

store.put(("user_123", "memories"), "1", {"text": "I love pizza"})
store.put(("user_123", "memories"), "2", {"text": "I am a plumber"})

items = store.search(
    ("user_123", "memories"), query="I'm hungry", limit=1
)


from langchain.embeddings import init_embeddings
from langchain.chat_models import init_chat_model
from langgraph.store.memory import InMemoryStore
from langgraph.graph import START, MessagesState, StateGraph
from langgraph.runtime import Runtime  

model = init_chat_model("gpt-5.4-mini")

# 创建启用语义搜索的存储
embeddings = init_embeddings("openai:text-embedding-3-small")
store = InMemoryStore(
    index={
        "embed": embeddings,
        "dims": 1536,
    }
)

store.put(("user_123", "memories"), "1", {"text": "I love pizza"})
store.put(("user_123", "memories"), "2", {"text": "I am a plumber"})

async def chat(state: MessagesState, runtime: Runtime):
    # 基于用户的最后一条消息进行搜索
    items = await runtime.store.asearch(
        ("user_123", "memories"), query=state["messages"][-1].content, limit=2
    )
    memories = "\n".join(item.value["text"] for item in items)
    memories = f"## Memories of user\n{memories}" if memories else ""
    response = await model.ainvoke(
        [
            {"role": "system", "content": f"You are a helpful assistant.\n{memories}"},
            *state["messages"],
        ]
    )
    return {"messages": [response]}


builder = StateGraph(MessagesState)
builder.add_node(chat)
builder.add_edge(START, "chat")
graph = builder.compile(store=store)

async for message, metadata in graph.astream(
    input={"messages": [{"role": "user", "content": "I'm hungry"}]},
    stream_mode="messages",
):
    print(message.content, end="")

​
管理短期记忆

启用短期记忆后，长时间的对话可能会超出 LLM 的上下文窗口。常见的解决方案有：
  • 修剪消息：移除前 N 条或后 N 条消息（在调用 LLM 之前）
  • 删除消息：从 LangGraph 状态中永久删除消息
  • 摘要消息：总结历史记录中较早的消息，并用摘要替换它们
  • 管理检查点：存储和检索消息历史记录
  • 自定义策略（例如，消息过滤等）
这使得智能体能够在不超出 LLM 上下文窗口的情况下跟踪对话。

​
修剪消息

大多数 LLM 都有一个最大支持的上下文窗口。一种决定何时截断消息的方法是：计算消息历史记录中的 token 数，并在接近该限制时进行截断。如果你使用的是 LangChain，可以使用 trim messages 工具，并指定要从列表中保留的 token 数量，以及处理边界时使用的 strategy（例如，保留最后 max_tokens 个 token）。 要修剪消息历史，请使用 trim_messages 函数： 
from langchain_core.messages.utils import (
    trim_messages,
    count_tokens_approximately  
)

def call_model(state: MessagesState):
    messages = trim_messages(
        state["messages"],
        strategy="last",
        token_counter=count_tokens_approximately,
        max_tokens=128,
        start_on="human",
        end_on=("human", "tool"),
    )
    response = model.invoke(messages)
    return {"messages": [response]}

builder = StateGraph(MessagesState)
builder.add_node(call_model)
...

from langchain_core.messages.utils import (
    trim_messages,
    count_tokens_approximately  
)
from langchain.chat_models import init_chat_model
from langgraph.graph import StateGraph, START, MessagesState

model = init_chat_model("claude-sonnet-4-6")
summarization_model = model.bind(max_tokens=128)

def call_model(state: MessagesState):
    messages = trim_messages(
        state["messages"],
        strategy="last",
        token_counter=count_tokens_approximately,
        max_tokens=128,
        start_on="human",
        end_on=("human", "tool"),
    )
    response = model.invoke(messages)
    return {"messages": [response]}

checkpointer = InMemorySaver()
builder = StateGraph(MessagesState)
builder.add_node(call_model)
builder.add_edge(START, "call_model")
graph = builder.compile(checkpointer=checkpointer)

config = {"configurable": {"thread_id": "1"}}
graph.invoke({"messages": "hi, my name is bob"}, config)
graph.invoke({"messages": "write a short poem about cats"}, config)
graph.invoke({"messages": "now do the same but for dogs"}, config)
final_response = graph.invoke({"messages": "what's my name?"}, config)

final_response["messages"][-1].pretty_print()

================================== 人工智能 消息 ==================================

你的名字是 Bob，这是你第一次自我介绍时提到的。

​
删除消息

你可以从图状态中删除消息来管理消息历史记录。这在你想移除特定消息或清空整个消息历史记录时很有用。 要从图状态中删除消息，可以使用 RemoveMessage。要使 RemoveMessage 生效，需要为状态键使用带有 add_messages reducer 的状态，例如 MessagesState。 要删除特定消息： 
from langchain.messages import RemoveMessage  

def delete_messages(state):
    messages = state["messages"]
    if len(messages) > 2:
        # 移除最早的两条消息
        return {"messages": [RemoveMessage(id=m.id) for m in messages[:2]]}
要删除所有消息： 
from langgraph.graph.message import REMOVE_ALL_MESSAGES  

def delete_messages(state):
    return {"messages": [RemoveMessage(id=REMOVE_ALL_MESSAGES)]}
删除消息时，务必确保 生成的消息历史是有效的。请检查你使用的 LLM 服务商的限制条件。例如：
  • 有些服务商要求消息历史以 user 消息开头。
  • 大多数服务商要求包含工具调用的 assistant 消息之后必须跟随相应的 tool 结果消息。
from langchain.messages import RemoveMessage  

def delete_messages(state):
    messages = state["messages"]
    if len(messages) > 2:
        # 移除最早的两条消息
        return {"messages": [RemoveMessage(id=m.id) for m in messages[:2]]}

def call_model(state: MessagesState):
    response = model.invoke(state["messages"])
    return {"messages": response}

builder = StateGraph(MessagesState)
builder.add_sequence([call_model, delete_messages])
builder.add_edge(START, "call_model")

checkpointer = InMemorySaver()
app = builder.compile(checkpointer=checkpointer)

for event in app.stream(
    {"messages": [{"role": "user", "content": "hi! I'm bob"}]},
    config,
    stream_mode="values"
):
    print([(message.type, message.content) for message in event["messages"]])

for event in app.stream(
    {"messages": [{"role": "user", "content": "what's my name?"}]},
    config,
    stream_mode="values"
):
    print([(message.type, message.content) for message in event["messages"]])

[('human', "hi! I'm bob")]
[('human', "hi! I'm bob"), ('ai', 'Hi Bob! How are you doing today? Is there anything I can help you with?')]
[('human', "hi! I'm bob"), ('ai', 'Hi Bob! How are you doing today? Is there anything I can help you with?'), ('human', "what's my name?")]
[('human', "hi! I'm bob"), ('ai', 'Hi Bob! How are you doing today? Is there anything I can help you with?'), ('human', "what's my name?"), ('ai', 'Your name is Bob.')]
[('human', "what's my name?"), ('ai', 'Your name is Bob.')]

​
摘要消息

如上所示，修剪或删除消息的问题在于，你可能会因从消息队列中剔除内容而丢失信息。因此，一些应用程序会受益于更复杂的方法：使用聊天模型来总结消息历史。 摘要 可以使用提示和编排逻辑来总结消息历史。例如，在 LangGraph 中，你可以扩展 MessagesState，使其包含一个 summary 键： 
from langgraph.graph import MessagesState
class State(MessagesState):
    summary: str
然后，你可以生成聊天历史的摘要，将任何现有摘要作为生成下一个摘要的上下文。这个 summarize_conversation 节点可以在 messages 状态键中积累了一定数量的消息后被调用。 
def summarize_conversation(state: State):

    # 首先，我们获取任何现有的摘要
    summary = state.get("summary", "")

    # 创建我们的摘要提示
    if summary:

        # 已经存在一个摘要
        summary_message = (
            f"This is a summary of the conversation to date: {summary}\n\n"
            "Extend the summary by taking into account the new messages above:"
        )

    else:
        summary_message = "Create a summary of the conversation above:"

    # 将提示添加到我们的历史记录中
    messages = state["messages"] + [HumanMessage(content=summary_message)]
    response = model.invoke(messages)

    # 删除除最近 2 条消息以外的所有消息
    delete_messages = [RemoveMessage(id=m.id) for m in state["messages"][:-2]]
    return {"summary": response.content, "messages": delete_messages}

from typing import Any, TypedDict

from langchain.chat_models import init_chat_model
from langchain.messages import AnyMessage
from langchain_core.messages.utils import count_tokens_approximately
from langgraph.graph import StateGraph, START, MessagesState
from langgraph.checkpoint.memory import InMemorySaver
from langmem.short_term import SummarizationNode, RunningSummary  

model = init_chat_model("claude-sonnet-4-6")
summarization_model = model.bind(max_tokens=128)

class State(MessagesState):
    context: dict[str, RunningSummary]

class LLMInputState(TypedDict):
    summarized_messages: list[AnyMessage]
    context: dict[str, RunningSummary]

summarization_node = SummarizationNode(
    token_counter=count_tokens_approximately,
    model=summarization_model,
    max_tokens=256,
    max_tokens_before_summary=256,
    max_summary_tokens=128,
)

def call_model(state: LLMInputState):
    response = model.invoke(state["summarized_messages"])
    return {"messages": [response]}

checkpointer = InMemorySaver()
builder = StateGraph(State)
builder.add_node(call_model)
builder.add_node("summarize", summarization_node)
builder.add_edge(START, "summarize")
builder.add_edge("summarize", "call_model")
graph = builder.compile(checkpointer=checkpointer)

# 调用图
config = {"configurable": {"thread_id": "1"}}
graph.invoke({"messages": "hi, my name is bob"}, config)
graph.invoke({"messages": "write a short poem about cats"}, config)
graph.invoke({"messages": "now do the same but for dogs"}, config)
final_response = graph.invoke({"messages": "what's my name?"}, config)

final_response["messages"][-1].pretty_print()
print("\n摘要:", final_response["context"]["running_summary"].summary)
  1. 我们将在 context 字段中追踪我们的动态摘要
（这正是 SummarizationNode 所要求的）。
  1. 定义将在节点间使用的私有状态，仅用于过滤
输入给 call_model 节点的内容。
  1. 我们在此传入一个私有的输入状态，以隔离由摘要节点返回的消息。
================================== 人工智能 消息 ==================================

From our conversation, I can see that you introduced yourself as Bob. That's the name you shared with me when we began talking.

摘要: In this conversation, I was introduced to Bob, who then asked me to write a poem about cats. I composed a poem titled "The Mystery of Cats" that captured cats' graceful movements, independent nature, and their special relationship with humans. Bob then requested a similar poem about dogs, so I wrote "The Joy of Dogs," which highlighted dogs' loyalty, enthusiasm, and loving companionship. Both poems were written in a similar style but emphasized the distinct characteristics that make each pet special.

​
管理检查点

你可以查看和删除由检查点存储器存储的信息。

​
查看线程状态

config = {
    "configurable": {
        "thread_id": "1",
        # 可选地，为特定检查点提供 ID，
        # 否则将显示最新的检查点
        # "checkpoint_id": "1f029ca3-1f5b-6704-8004-820c16b69a5a"  #

    }
}
graph.get_state(config)

StateSnapshot(
    values={'messages': [HumanMessage(content="hi! I'm bob"), AIMessage(content='Hi Bob! How are you doing today?), HumanMessage(content="what's my name?"), AIMessage(content='Your name is Bob.')]}, next=(),
    config={'configurable': {'thread_id': '1', 'checkpoint_ns': '', 'checkpoint_id': '1f029ca3-1f5b-6704-8004-820c16b69a5a'}},
    metadata={
        'source': 'loop',
        'writes': {'call_model': {'messages': AIMessage(content='Your name is Bob.')}},
        'step': 4,
        'parents': {},
        'thread_id': '1'
    },
    created_at='2025-05-05T16:01:24.680462+00:00',
    parent_config={'configurable': {'thread_id': '1', 'checkpoint_ns': '', 'checkpoint_id': '1f029ca3-1790-6b0a-8003-baf965b6a38f'}},
    tasks=(),
    interrupts=()
)

​
查看线程历史

config = {
    "configurable": {
        "thread_id": "1"
    }
}
list(graph.get_state_history(config))

[
    StateSnapshot(
        values={'messages': [HumanMessage(content="hi! I'm bob"), AIMessage(content='Hi Bob! How are you doing today? Is there anything I can help you with?'), HumanMessage(content="what's my name?"), AIMessage(content='Your name is Bob.')]},
        next=(),
        config={'configurable': {'thread_id': '1', 'checkpoint_ns': '', 'checkpoint_id': '1f029ca3-1f5b-6704-8004-820c16b69a5a'}},
        metadata={'source': 'loop', 'writes': {'call_model': {'messages': AIMessage(content='Your name is Bob.')}}, 'step': 4, 'parents': {}, 'thread_id': '1'},
        created_at='2025-05-05T16:01:24.680462+00:00',
        parent_config={'configurable': {'thread_id': '1', 'checkpoint_ns': '', 'checkpoint_id': '1f029ca3-1790-6b0a-8003-baf965b6a38f'}},
        tasks=(),
        interrupts=()
    ),
    StateSnapshot(
        values={'messages': [HumanMessage(content="hi! I'm bob"), AIMessage(content='Hi Bob! How are you doing today? Is there anything I can help you with?'), HumanMessage(content="what's my name?")]},
        next=('call_model',),
        config={'configurable': {'thread_id': '1', 'checkpoint_ns': '', 'checkpoint_id': '1f029ca3-1790-6b0a-8003-baf965b6a38f'}},
        metadata={'source': 'loop', 'writes': None, 'step': 3, 'parents': {}, 'thread_id': '1'},
        created_at='2025-05-05T16:01:23.863421+00:00',
        parent_config={...}
        tasks=(PregelTask(id='8ab4155e-6b15-b885-9ce5-bed69a2c305c', name='call_model', path=('__pregel_pull', 'call_model'), error=None, interrupts=(), state=None, result={'messages': AIMessage(content='Your name is Bob.')}),),
        interrupts=()
    ),
    StateSnapshot(
        values={'messages': [HumanMessage(content="hi! I'm bob"), AIMessage(content='Hi Bob! How are you doing today? Is there anything I can help you with?')]},
        next=('__start__',),
        config={...},
        metadata={'source': 'input', 'writes': {'__start__': {'messages': [{'role': 'user', 'content': "what's my name?"}]}}, 'step': 2, 'parents': {}, 'thread_id': '1'},
        created_at='2025-05-05T16:01:23.863173+00:00',
        parent_config={...}
        tasks=(PregelTask(id='24ba39d6-6db1-4c9b-f4c5-682aeaf38dcd', name='__start__', path=('__pregel_pull', '__start__'), error=None, interrupts=(), state=None, result={'messages': [{'role': 'user', 'content': "what's my name?"}]}),),
        interrupts=()
    ),
    StateSnapshot(
        values={'messages': [HumanMessage(content="hi! I'm bob"), AIMessage(content='Hi Bob! How are you doing today? Is there anything I can help you with?')]},
        next=(),
        config={...},
        metadata={'source': 'loop', 'writes': {'call_model': {'messages': AIMessage(content='Hi Bob! How are you doing today? Is there anything I can help you with?')}}, 'step': 1, 'parents': {}, 'thread_id': '1'},
        created_at='2025-05-05T16:01:23.862295+00:00',
        parent_config={...}
        tasks=(),
        interrupts=()
    ),
    StateSnapshot(
        values={'messages': [HumanMessage(content="hi! I'm bob")]},
        next=('call_model',),
        config={...},
        metadata={'source': 'loop', 'writes': None, 'step': 0, 'parents': {}, 'thread_id': '1'},
        created_at='2025-05-05T16:01:22.278960+00:00',
        parent_config={...}
        tasks=(PregelTask(id='8cbd75e0-3720-b056-04f7-71ac805140a0', name='call_model', path=('__pregel_pull', 'call_model'), error=None, interrupts=(), state=None, result={'messages': AIMessage(content='Hi Bob! How are you doing today? Is there anything I can help you with?')}),),
        interrupts=()
    ),
    StateSnapshot(
        values={'messages': []},
        next=('__start__',),
        config={'configurable': {'thread_id': '1', 'checkpoint_ns': '', 'checkpoint_id': '1f029ca3-0870-6ce2-bfff-1f3f14c3e565'}},
        metadata={'source': 'input', 'writes': {'__start__': {'messages': [{'role': 'user', 'content': "hi! I'm bob"}]}}, 'step': -1, 'parents': {}, 'thread_id': '1'},
        created_at='2025-05-05T16:01:22.277497+00:00',
        parent_config=None,
        tasks=(PregelTask(id='d458367b-8265-812c-18e2-33001d199ce6', name='__start__', path=('__pregel_pull', '__start__'), error=None, interrupts=(), state=None, result={'messages': [{'role': 'user', 'content': "hi! I'm bob"}]}),),
        interrupts=()
    )
]

​
删除线程的所有检查点

thread_id = "1"
checkpointer.delete_thread(thread_id)

​
数据库管理

如果你使用任何由数据库支持的持久化实现（例如 Postgres、Redis 或 Oracle）来存储短期和/或长期记忆，则需要在你将其用于数据库之前运行迁移以设置所需的模式。 按照惯例，大多数特定于数据库的库会在检查点存储器或存储实例上定义一个 setup() 方法，用于运行所需的迁移。但是，你应该核对你的 BaseCheckpointSaver 或 BaseStore 的具体实现，以确认确切的方法名称和用法。 我们建议将迁移作为一个专门的部署步骤来运行，或者确保它们作为服务器启动的一部分被运行。
Connect these docs to Claude, VSCode, and more via MCP for real-time answers.
Edit this page on GitHub or file an issue.