langchain-ai/langgraph
1
0
Fork 0
mirror of https://github.com/langchain-ai/langgraph.git synced 2026-07-15 12:35:43 -04:00
Code Issues 314 Packages Projects Releases 17 Wiki Activity

[PR #2556] Add MongoDB Checkpoint Saver Implementation for LangGraph #2777

New Issue
Closed
opened 2026-02-20 17:47:33 -05:00 by yindo · 0 comments
yindo commented 2026-02-20 17:47:33 -05:00
Owner
Copy Link
Copy Source

Original Pull Request: https://github.com/langchain-ai/langgraph/pull/2556

State: closed
Merged: No

Pull Request Description:

This PR introduces a MongoDB-based checkpoint saver implementation for LangGraph. The implementation includes both synchronous and asynchronous versions, which can be used to save and retrieve checkpoints in a MongoDB database.

Key Changes:

  • AsyncMongoDBSaver: An asynchronous implementation of the checkpoint saver using MongoDB, designed for high performance and non-blocking operations.
  • MongoDBSaver: A synchronous implementation of the checkpoint saver using MongoDB, for simpler use cases where asynchronous operations are not necessary.
  • Example Usage: Added examples for using the MongoDB checkpoint savers with LangGraph agents, demonstrating how to use both the synchronous and asynchronous checkpointing methods.

Features:

  • Supports MongoDB version 4.x and later.
  • Utilizes the motor library for asynchronous MongoDB operations.
  • Provides a simple, configuration-driven API for connecting to a MongoDB database and saving/retrieving checkpoints.

Example Code:

  1. Synchronous Checkpoint Saver:
    from typing import Literal
    from langchain_core.runnables import ConfigurableField
    from langchain_core.tools import tool
    from langchain_openai import ChatOpenAI
    from langgraph.checkpoint.mongodb import MongoDBSaver
    from langgraph.checkpoint.mongodb.aio import AsyncMongoDBSaver
    from langgraph.prebuilt import create_react_agent

    @tool
    def get_weather(city: Literal["nyc", "sf"]):
        if city == "nyc":
            return "It might be cloudy in nyc"
        elif city == "sf":
            return "It's always sunny in sf"
        else:
            raise AssertionError("Unknown city")


    tools = [get_weather]  # List of tools to be used by the agent
    model = ChatOpenAI(model_name="gpt-4o-mini", temperature=0, api_key="your_api_key")

    # Create a MongoDBSaver instance and use it as a checkpointer
    with MongoDBSaver.from_conn_info(
        url="mongodb://localhost:27017", db_name="checkpoints"
    ) as checkpointer:

        # Create a React agent using the model and tools, along with the MongoDB checkpointer
        graph = create_react_agent(model, tools=tools, checkpointer=checkpointer)

        # Example configuration for the agent's execution
        config = {"configurable": {"thread_id": "1"}}

        # Invoke the agent with a query
        res = graph.invoke({"messages": [("human", "what's the weather in sf")]}, config)

        # Retrieve the latest checkpoint
        latest_checkpoint = checkpointer.get(config)
        latest_checkpoint_tuple = checkpointer.get_tuple(config)

        # List all checkpoint tuples
        checkpoint_tuples = list(checkpointer.list(config))
  1. Asynchronous Checkpoint Saver: 
     from typing import Literal
 from langchain_core.tools import tool
 from langchain_openai import ChatOpenAI
 from langgraph.checkpoint.mongodb.aio import AsyncMongoDBSaver
 from langgraph.prebuilt import create_react_agent

 @tool
 def get_weather(city: Literal["nyc", "sf"]):
     if city == "nyc":
         return "It might be cloudy in nyc"
     elif city == "sf":
         return "It's always sunny in sf"
     else:
         raise AssertionError("Unknown city")


 tools = [get_weather]  # List of tools to be used by the agent
 model = ChatOpenAI(model_name="gpt-4o-mini", temperature=0, api_key="your_api_key")

 # Create a AsyncMongoDBSaver instance and use it as a checkpointer
 async with AsyncMongoDBSaver.from_conn_info(
     url="mongodb://localhost:27017", db_name="checkpoints"
 ) as checkpointer:

     # Create a React agent using the model and tools, along with the MongoDB checkpointer
     graph = create_react_agent(model, tools=tools, checkpointer=checkpointer)

     # Example configuration for the agent's execution
     config = {"configurable": {"thread_id": "1"}}

     # ainvoke the agent with a query
     res = await graph.ainvoke({"messages": [("human", "what's the weather in sf")]}, config)

     # Retrieve the latest checkpoint
     latest_checkpoint = checkpointer.get(config)
     latest_checkpoint_tuple = checkpointer.get_tuple(config)

     # List all checkpoint tuples
     checkpoint_tuples = list(checkpointer.list(config))

Checklist:

  • Code implemented and tested
  • Documentation updated (README.md, usage examples)
  • Tests added
  • Code linting and formatting checks
**Original Pull Request:** https://github.com/langchain-ai/langgraph/pull/2556 **State:** closed **Merged:** No --- ### Pull Request Description: This PR introduces a MongoDB-based checkpoint saver implementation for LangGraph. The implementation includes both synchronous and asynchronous versions, which can be used to save and retrieve checkpoints in a MongoDB database. #### Key Changes: - **`AsyncMongoDBSaver`**: An asynchronous implementation of the checkpoint saver using MongoDB, designed for high performance and non-blocking operations. - **`MongoDBSaver`**: A synchronous implementation of the checkpoint saver using MongoDB, for simpler use cases where asynchronous operations are not necessary. - **Example Usage**: Added examples for using the MongoDB checkpoint savers with LangGraph agents, demonstrating how to use both the synchronous and asynchronous checkpointing methods. #### Features: - Supports MongoDB version 4.x and later. - Utilizes the `motor` library for asynchronous MongoDB operations. - Provides a simple, configuration-driven API for connecting to a MongoDB database and saving/retrieving checkpoints. #### Example Code: 1. **Synchronous Checkpoint Saver**: ```python from typing import Literal from langchain_core.runnables import ConfigurableField from langchain_core.tools import tool from langchain_openai import ChatOpenAI from langgraph.checkpoint.mongodb import MongoDBSaver from langgraph.checkpoint.mongodb.aio import AsyncMongoDBSaver from langgraph.prebuilt import create_react_agent @tool def get_weather(city: Literal["nyc", "sf"]): if city == "nyc": return "It might be cloudy in nyc" elif city == "sf": return "It's always sunny in sf" else: raise AssertionError("Unknown city") tools = [get_weather] # List of tools to be used by the agent model = ChatOpenAI(model_name="gpt-4o-mini", temperature=0, api_key="your_api_key") # Create a MongoDBSaver instance and use it as a checkpointer with MongoDBSaver.from_conn_info( url="mongodb://localhost:27017", db_name="checkpoints" ) as checkpointer: # Create a React agent using the model and tools, along with the MongoDB checkpointer graph = create_react_agent(model, tools=tools, checkpointer=checkpointer) # Example configuration for the agent's execution config = {"configurable": {"thread_id": "1"}} # Invoke the agent with a query res = graph.invoke({"messages": [("human", "what's the weather in sf")]}, config) # Retrieve the latest checkpoint latest_checkpoint = checkpointer.get(config) latest_checkpoint_tuple = checkpointer.get_tuple(config) # List all checkpoint tuples checkpoint_tuples = list(checkpointer.list(config)) ``` 2. **Asynchronous Checkpoint Saver**: ```python from typing import Literal from langchain_core.tools import tool from langchain_openai import ChatOpenAI from langgraph.checkpoint.mongodb.aio import AsyncMongoDBSaver from langgraph.prebuilt import create_react_agent @tool def get_weather(city: Literal["nyc", "sf"]): if city == "nyc": return "It might be cloudy in nyc" elif city == "sf": return "It's always sunny in sf" else: raise AssertionError("Unknown city") tools = [get_weather] # List of tools to be used by the agent model = ChatOpenAI(model_name="gpt-4o-mini", temperature=0, api_key="your_api_key") # Create a AsyncMongoDBSaver instance and use it as a checkpointer async with AsyncMongoDBSaver.from_conn_info( url="mongodb://localhost:27017", db_name="checkpoints" ) as checkpointer: # Create a React agent using the model and tools, along with the MongoDB checkpointer graph = create_react_agent(model, tools=tools, checkpointer=checkpointer) # Example configuration for the agent's execution config = {"configurable": {"thread_id": "1"}} # ainvoke the agent with a query res = await graph.ainvoke({"messages": [("human", "what's the weather in sf")]}, config) # Retrieve the latest checkpoint latest_checkpoint = checkpointer.get(config) latest_checkpoint_tuple = checkpointer.get_tuple(config) # List all checkpoint tuples checkpoint_tuples = list(checkpointer.list(config)) ``` #### Checklist: - [x] Code implemented and tested - [x] Documentation updated (README.md, usage examples) - [x] Tests added - [x] Code linting and formatting checks
yindo added the pull-request label 2026-02-20 17:47:33 -05:00
yindo closed this issue 2026-02-20 17:47:33 -05:00
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
agents
awaiting author revision
bug
change
chat models
checkpointer
ci
code quality
code quality
dependencies
documentation
drawing
duplicate
enhancement
good first issue
help wanted
invalid
maintainer
meta
missing information
not maintained by langchain
open-swe-auto
open-swe-max
open-swe-max-auto
pending
performance
pull-request
Mirrored from GitHub Pull Request
 question
requires design
sdk
server
stale
studio
tests
type checking
usage question
v0.5
v1.1
No Label pull-request
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
yindo
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: langchain-ai/langgraph#2777
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?