langgraph-checkpoint-aws

What it is and what it does

LangGraph Checkpoint AWS is a persistence layer for LangGraph agents that integrates with AWS services. It offers three main storage strategies: Bedrock AgentCore Memory for managed session storage, DynamoDB with optional S3 offloading for checkpoint and document storage with vector search, and Valkey (Redis-compatible) for high-performance caching and checkpointing. The package acts as a checkpointer and document store, allowing LangGraph agents to resume from saved state across sessions and to retrieve long-term memories and preferences.

The package is designed for developers building stateful, resumable agents on AWS infrastructure. It handles the integration between LangGraph's checkpoint interface and AWS backend services, including automatic compression, TTL management, and intelligent offloading of large checkpoints to S3. Runtime dependencies are langgraph, langgraph-checkpoint, boto3, and typing_extensions; Valkey support is optional.

Use it for:

• Build resumable chatbot agents that persist conversation state across sessions using Bedrock AgentCore Memory.
• Store agent checkpoints in DynamoDB with automatic S3 offloading for long-running or memory-intensive workflows.
• Cache LLM responses and computation results using Valkey to reduce latency and API costs.
• Implement semantic search over agent memories and preferences stored in DynamoDB vector indexes.
• Deploy multi-actor agent systems where each actor maintains separate session state via configurable thread and actor IDs.

Worth the install?

AI-flagged interpretation of the facts on this page — verify before relying

Provides AWS-based persistence backends for LangGraph agents, including Bedrock AgentCore Memory, DynamoDB with S3 offloading, and Valkey checkpoint and document storage.

Yes. The package is actively maintained, has low install friction, carries a permissive MIT license, and provides well-integrated AWS persistence options for LangGraph agents. Choose it if you are building stateful agents on AWS and need checkpoint storage, document retrieval, or LLM response caching. No security vulnerabilities are known.

Install

langgraph-checkpoint-aws on PyPI

pip install langgraph-checkpoint-aws

uv add langgraph-checkpoint-aws

poetry add langgraph-checkpoint-aws

Installing langgraph-checkpoint-aws

Before you install

Low friction install with four runtime dependencies. Actively maintained as of 2026-08-14 with recent releases; repository shows 340 stars and no archived status.

License in practice

MIT license permits commercial and private use with minimal restrictions; you must include a copy of the license and copyright notice.

Quickstart

pip install langgraph-checkpoint-aws

from langgraph_checkpoint_aws import AgentCoreMemorySaver
from langgraph.prebuilt import create_react_agent

checkpointer = AgentCoreMemorySaver("YOUR_MEMORY_ID", region_name="us-west-2")
graph = create_react_agent(model=model, tools=tools, checkpointer=checkpointer)
config = {"configurable": {"thread_id": "session-1", "actor_id": "agent-1"}}
response = graph.invoke({"messages": [("human", "your prompt")]}, config=config)

Requires Python 3.10 or later and AWS credentials configured for the target region.

Verify before relying

• Whether DynamoDB vector search via semantic indexes requires additional setup or AWS service enablement beyond standard DynamoDB.
• Performance characteristics and latency profiles for each storage backend under production load.
• Exact compression algorithm and storage overhead reduction achieved by enable_checkpoint_compression.
• Whether Valkey optional dependency is required for all use cases or only for specific storage backends.

Package facts

Evidence: langgraph_checkpoint_aws-1.2.2-py3-none-any.whl

Keywords: aws, bedrock, langchain, langgraph, checkpointer, dynamodb, elasticache, valkey

Tags

More Application Frameworks packages