langgraph-checkpoint-redis

What it is and what it does

langgraph-checkpoint-redis is a Redis backend for LangGraph's checkpoint and store systems, allowing you to persist agent state and conversation history in Redis instead of memory or local storage. It provides both standard and shallow checkpoint savers (sync and async), plus key-value stores with optional vector search for semantic caching and tool caching in LangChain agents.

The package wraps redis and redisvl to manage checkpoint serialization, TTL-based expiration, and index creation. It handles the complexity of connecting to different Redis deployments—standard Redis, Azure Managed Redis, Redis Enterprise—by detecting cluster mode and configuring clients appropriately. You call .setup() once to initialize indices, then use put/get/list methods to store and retrieve checkpoints keyed by thread_id and namespace.

Use it for:

• Store LangGraph agent execution checkpoints in Redis for multi-turn conversation recovery across server restarts.
• Implement semantic caching for LLM tool calls using Redis vector search to avoid redundant API calls.
• Scale agent state across multiple worker processes by persisting checkpoints to a shared Redis instance.
• Expire old conversation history automatically using Redis TTL to manage storage costs.
• Build agent middleware for conversation memory and tool result caching in production deployments.

Worth the install?

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

Provides Redis-backed checkpoint storage and key-value stores for LangGraph agents, with optional vector search capabilities via RedisJSON and RediSearch modules.

Yes, if you are building LangGraph agents and need persistent, scalable checkpoint storage. The package is actively maintained, has low install friction, and solves a real problem for production agent deployments. The main gotcha is the Redis module requirement (RedisJSON, RediSearch)—verify your Redis deployment includes these before installing, especially on Azure or Redis Enterprise.

Install

langgraph-checkpoint-redis on PyPI

pip install langgraph-checkpoint-redis

uv add langgraph-checkpoint-redis

poetry add langgraph-checkpoint-redis

Installing langgraph-checkpoint-redis

Before you install

Low install friction; pure Python wheel with straightforward dependencies. Active maintenance (last commit 2026-07-14, 31 days since release). Requires Redis 8.0+ with RedisJSON and RediSearch built-in, or Redis Stack for older versions—this is a runtime requirement, not a Python install issue.

License in practice

MIT license (permissive); you can use, modify, and distribute this package freely in commercial and open-source projects with minimal restrictions.

Quickstart

from langgraph.checkpoint.redis import RedisSaver

with RedisSaver.from_conn_string("redis://localhost:6379") as checkpointer:
    checkpointer.setup()
    checkpointer.put({"configurable": {"thread_id": "1"}}, checkpoint_data, {}, {})

Requires Redis 8.0+ (or Redis Stack for older versions) with RedisJSON and RediSearch modules enabled; Azure Managed Redis and Redis Enterprise require standard Redis client, not cluster-aware client.

Verify before relying

• Whether redisvl>=0.5.1 and orjson are used directly by the package or are transitive dependencies only.
• Performance characteristics (throughput, latency) for checkpoint read/write operations at scale.
• Whether the shallow checkpoint implementations (ShallowRedisSaver) are production-ready or experimental.

Package facts

Evidence: langgraph_checkpoint_redis-0.5.1-py3-none-any.whl

Keywords: ai, redis, redis-client, vector-database, agents, langgraph, langchain

Tags

More Artificial Intelligence packages