Skip to main content
LangSmith Cloud is a managed platform for deploying your agents. LangChain hosts and operates the control plane, data plane, Agent Server runtime, and supporting databases on AWS and GCP. Push code to a connected GitHub repository or invoke the langgraph deploy CLI, and the platform handles build, provisioning, scaling, and ongoing operations. Deployments come in two types:
  • Serverless: Scales to zero when idle. Suits background agents, testing, and latency-tolerant production workloads.
  • Dedicated: Stays on and autoscales across replicas. Suits latency-sensitive production workloads.
Agent deployments running on Cloud require a Plus plan or above. Before creating your first agent deployment, verify that your application runs locally with langgraph dev. Refer to Local development and testing.

Deploy on Cloud

Step-by-step setup guide for creating, configuring, and managing Cloud deployments from the LangSmith UI or the langgraph deploy CLI.

Cloud platform features

Reference for Cloud-only platform behavior: data regions, static IPs, payload limits, deployment types, and managed database provisioning.

Quickstart

Deploy your first LangGraph application to Cloud in a few minutes.
To deploy a code-first Deep Agent without standing up your own Agent Server, Managed Deep Agents offers a CLI-first managed runtime in public beta.

​
Next steps

Run the quickstart

Deploy a starter LangGraph application end-to-end.

Read the full deploy guide

Configure environment variables, secrets, revisions, and deployment settings.
Connect these docs to your agent of choice via MCP for real-time answers.
Edit this page on GitHub or file an issue.