LangGraph Platform is now generally available. Build, deploy, and manage long-running, stateful agents — at scale.

LangGraph Platform is our purpose-built infrastructure for deploying and scaling long-running, stateful agents. Since our beta, hundreds of teams have used LangGraph Platform to deploy their agents into production. Now it’s generally available for all.

What LangGraph Platform includes:

Agent infrastructure is hard. Many agents are long-running, rely on async collaboration, and bursty.

LangGraph Platform makes it easy to deploy your agents, to accelerate your agent development using visual workflows, and to centrally manage your agents across the organization. It includes:

• 1-click deploy from GitHub → live in minutes

• Built-in memory & persistence for async, long-running agents

• Scalable API endpoints for any interaction pattern

• LangGraph Studio for visual debugging, retries, and rewinds

• Agent registry to manage and version agents across your org

Deployment options for LangGraph Platform

• Cloud (SaaS): Fastest way to get started, fully managed and easy deployment from within LangSmith. Available on our Plus plan and Enterprise plan. To get started, read the docs (Plus plan) or contact sales (Enterprise).

• Hybrid: SaaS control plane with self-hosted data plane — ideal for teams with sensitive data but still want more of a managed service. Available only on the Enterprise plan. To get started, contact sales

• Fully Self-Hosted: Run the entire platform within your own infrastructure. No data leaves your VPC. Available on the Enterprise plan. To get started, contact sales.

• If you want to try out a basic version of our LangGraph server in your environment, you can also self-host on our Developer plan and get up to 100k nodes executed per month for free – great to give you exposure to LangGraph Platform and run hobbyist projects on your infra. To get started on the Developer tier, read the docs.

Learn more in our blog post here: https://blog.langchain.dev/langgraph-platform-ga/