langgraph-checkpoint
Library with base interfaces for LangGraph checkpoint savers.
Decision gist · record as of 2026-08-14
Yes. This is a foundational library for any LangGraph application requiring state persistence, multi-threaded execution, or resumable workflows. Low install friction, active maintenance, no known vulnerabilities, and MIT licensing make it a safe dependency. Install it if you are building LangGraph applications; it is a required abstraction layer for production deployments.AI-flagged interpretation of the facts on this page — verify before relying
Before you install
- Requires Python 3.10 or later.
- Deserialization security: set LANGGRAPH_STRICT_MSGPACK=true or pass allowed_msgpack_modules to JsonPlusSerializer to restrict types in production.
- Low friction: pure Python wheel with only two runtime dependencies (langchain-core and ormsgpack).
License · maintenance · safety
MIT (permissive) — MIT license permits unrestricted use, modification, and distribution with minimal legal friction for commercial or private projects.
last release 2026-08-07 (7 days) · last repo commit 2026-08-13 · 39,639 stars
0 known vulnerabilities (OSV.dev, 2026-08-14) · 55,119,367 downloads/mo, #539 on PyPI
Alternatives
Verify before relying
pip install langgraph-checkpoint
from langgraph.checkpoint.memory import InMemorySaver
checkpointer = InMemorySaver()
write_config = {"configurable": {"thread_id": "1", "checkpoint_ns": ""}}
checkpointer.put(write_config, checkpoint_data, {}, {})
checkpoint = checkpointer.get(write_config)- Whether the package includes concrete implementations beyond InMemorySaver (e.g., SQL or cloud-backed savers) or only the interface.
- Performance characteristics and scalability limits for checkpoint storage and retrieval under load.
- Compatibility guarantees with specific LangGraph versions beyond the langchain-core dependency.
What it is and what it does
langgraph-checkpoint defines the base interface and serialization protocol that LangGraph checkpointers must implement to persist graph state. A checkpointer saves snapshots of graph execution at each superstep, identified by a thread_id and optional checkpoint_id, enabling workflows to resume from any saved point. This is essential for human-in-the-loop interactions, maintaining state across multiple user conversations in multi-tenant applications, and recovering from mid-execution failures.
The package provides BaseCheckpointSaver as the interface contract, specifying methods like put, get_tuple, list, and delete_thread for both synchronous and asynchronous execution. It also includes JsonPlusSerializer, a default serialization handler that converts Python objects—including LangChain primitives, datetimes, and enums—to and from msgpack format. By default the deserializer accepts any type found in checkpoint data; production deployments should enable LANGGRAPH_STRICT_MSGPACK or supply an allowed_msgpack_modules list to restrict deserialization to known-safe types.
Use it for
- Build multi-tenant chat applications where each user conversation is isolated in its own thread with independent state history.
- Implement human-in-the-loop workflows where graph execution can be paused, reviewed, and resumed from a specific checkpoint.
- Create durable LLM pipelines that survive node failures by storing pending writes and replaying only incomplete nodes on recovery.
- Enable conversation memory and context persistence across multiple invocations of the same LangGraph application.
- Develop debugging and replay tools that load and re-execute a graph from any saved checkpoint in its execution history.
Worth the install?
AI-flagged interpretation of the facts on this page. Verify before relying on it.
Yes.
This is a foundational library for any LangGraph application requiring state persistence, multi-threaded execution, or resumable workflows. Low install friction, active maintenance, no known vulnerabilities, and MIT licensing make it a safe dependency. Install it if you are building LangGraph applications; it is a required abstraction layer for production deployments.
Install
langgraph-checkpoint on PyPI
Before you install
Low friction: pure Python wheel with only two runtime dependencies (langchain-core and ormsgpack). Active maintenance with a release 7 days ago and 39639 repository stars.
Requires Python 3.10 or later. Deserialization security: set LANGGRAPH_STRICT_MSGPACK=true or pass allowed_msgpack_modules to JsonPlusSerializer to restrict types in production.
License in practice
MIT license permits unrestricted use, modification, and distribution with minimal legal friction for commercial or private projects.
Quickstart
pip install langgraph-checkpoint
from langgraph.checkpoint.memory import InMemorySaver
checkpointer = InMemorySaver()
write_config = {"configurable": {"thread_id": "1", "checkpoint_ns": ""}}
checkpointer.put(write_config, checkpoint_data, {}, {})
checkpoint = checkpointer.get(write_config)Verify before relying
- Whether the package includes concrete implementations beyond InMemorySaver (e.g., SQL or cloud-backed savers) or only the interface.
- Performance characteristics and scalability limits for checkpoint storage and retrieval under load.
- Compatibility guarantees with specific LangGraph versions beyond the langchain-core dependency.
Package facts
| License | MIT permissive |
| Python support | Supports the current Python release >=3.10 |
| Install friction | Low. Pure-Python wheel |
| Runtime dependencies | 2 packageslangchain-coreormsgpack |
| Maintenance | Actively maintained 7 days since the last release |
| Last repo commit | |
| First released | |
| Downloads | 55,119,367 / month, #539 on PyPI 30-day window, as of 2026-08-14 |
| Known vulnerabilities | None known OSV.dev, checked 2026-08-14 |
Evidence: langgraph_checkpoint-4.2.0-py3-none-any.whl
Tags
Let your AI agent find packages like this
Example. Real query, live index.
You found this page by searching. An agent finds it by wishing: SkillFed indexes 14,416 PyPI packages by what they can do, searchable in plain language.
wish › “graph execution resumption”
- langgraph-checkpointProvides the base interface and serialization protocol for LangGraph…
- langgraph-checkpoint-mongodbProvides a MongoDB-backed checkpoint storage implementation for…
- pymgclientpymgclient is a Python adapter for Memgraph graph database,…
Give your agent the search over MCP, or paste the wish link into any chat.
Similar packages
Provides a Postgres-backed checkpoint saver for LangGraph, enabling durable state persistence for long-running workflows and agents.
Provides a SQLite-backed checkpoint saver for LangGraph, enabling state persistence in LLM applications with both sync and async support.
Install it if you are building LangGraph applications and need local or lightweight checkpoint storage; skip it if you require distributed state management or are…
Provides AWS-based persistence backends for LangGraph agents, including Bedrock AgentCore Memory, DynamoDB with S3 offloading, and Valkey checkpoint and document storage.
Provides Redis-backed checkpoint storage and key-value stores for LangGraph agents, with optional vector search capabilities via RedisJSON and RediSearch modules.
The main gotcha is the Redis module requirement (RedisJSON, RediSearch)—verify your Redis deployment includes these before installing, especially on Azure or Redis…
Provides a MongoDB-backed checkpoint storage implementation for LangGraph, enabling state persistence and recovery in agentic workflows.
Provides an in-memory runtime implementation for the LangGraph API server, enabling graph-based workflow execution without external persistence.
However, clarify the Elastic-2.0 license terms for your use case first, and confirm that in-memory execution meets your durability and scaling requirements before…
See also langsmith-fetch · langgraph · langgraph-cli · orbax-checkpoint