Deploy your app to cloud

This quickstart shows you how to deploy an application to LangSmith Cloud in GCP using the langgraph deploy command. Any app that exports a graph from a langgraph.json config deploys the same way, regardless of which framework you used to author the agent.

For a comprehensive Cloud deployment guide including GitHub-based deployments and all configuration options, refer to the Cloud deployment setup guide.

The langgraph deploy command is in beta. On LangSmith Cloud (SaaS) in AWS, local builds require langgraph-cli 0.4.35 or later.

Prerequisites

Before you begin, ensure you have:

A LangSmith account on the Plus plan or above and an API key.
(Optional) Docker installed and the Docker daemon running for local builds. Not required for remote builds. Install Docker Desktop. If Docker is not available, langgraph deploy triggers a remote build automatically.
(Optional) On Apple Silicon (M1/M2/M3): Docker Buildx for cross-compiling to linux/amd64 during local builds.
The LangGraph CLI:
uv tool install langgraph-cli

Get started with your coding agent

A coding agent can follow this quickstart to prepare, test, and deploy your application. To get started, paste this prompt into your coding agent:

Deploy an application to LangSmith Cloud

To follow the steps yourself, continue below.

1. Create a deployable app

langgraph deploy deploys any project whose langgraph.json exports a graph. Pick the path that matches how you author your agent:

LangGraph template
Bring your own framework

Create a new app from the new-langgraph-project-python template:

langgraph new path/to/your/app --template new-langgraph-project-python
cd path/to/your/app

Run langgraph new without --template for an interactive menu of available templates.

Agents authored with Claude Agent SDK, Strands, CrewAI, AutoGen, or Google ADK deploy through the same CLI once they expose a graph from langgraph.json. For end-to-end examples, see Deploy other frameworks. Once your project exports a graph, return here for the remaining steps.

2. Set your API key

Add your LangSmith API key to a .env file in your project root:

LANGSMITH_API_KEY=lsv2_...

The langgraph deploy command reads this automatically. Alternatively, pass it inline:

LANGSMITH_API_KEY=lsv2_... langgraph deploy

3. Deploy

Deploy directly from the CLI or via the UI.

Deploy from CLI
Deploy from Studio

Run the deploy command from your project directory:

langgraph deploy

This creates a Serverless Agent Server named after your project directory. Use --name to change the name, or --agent-server-type dedicated for a Dedicated Agent Server.

To update an existing Agent Server after making code changes, re-run langgraph deploy. It finds the existing Agent Server by name and updates it in place.

You can also use langgraph deploy list to see all Agent Servers, langgraph deploy logs to tail runtime logs, and langgraph deploy delete <ID> to remove an Agent Server. For details, refer to the CLI reference.

To deploy from studio:

Start the local development server. This will automatically open up Studio, an interactive agent IDE.

langgraph dev

Click the deploy button.

4. Test in Studio

Studio is an interactive agent IDE connected directly to your Agent Server. Use it to send messages, inspect intermediate state at each node, edit state mid-run, and replay from any prior checkpoint without writing code. Once the Agent Server is ready:

Go to LangSmith and select Agent Servers in the left sidebar.
Select your Agent Server to view its details.
Click Studio in the top right corner to open Studio.

5. Test the API

In LangSmith, copy the Agent Server address from the URL row in the Agent Server details view. Use it to call your application:

Python SDK (Async)
Python SDK (Sync)
JavaScript SDK
Rest API

Install the LangGraph Python SDK:
pip install langgraph-sdk

Send a message to the assistant (stateless run):

from langgraph_sdk import get_client

client = get_client(url="your-deployment-url", api_key="your-langsmith-api-key")

async for chunk in client.runs.stream(
    None,  # Threadless run
    "agent", # Name of assistant. Defined in langgraph.json.
    input={
        "messages": [{
            "role": "human",
            "content": "Say hello.",
        }],
    },
    stream_mode="updates",
):
    print(f"Receiving new event of type: {chunk.event}...")
    print(chunk.data)
    print("\n\n")

Install the LangGraph Python SDK:
pip install langgraph-sdk

Send a message to the assistant (threadless run):

from langgraph_sdk import get_sync_client

client = get_sync_client(url="your-deployment-url", api_key="your-langsmith-api-key")

for chunk in client.runs.stream(
    None,  # Threadless run
    "agent", # Name of assistant. Defined in langgraph.json.
    input={
        "messages": [{
            "role": "human",
            "content": "Say hello.",
        }],
    },
    stream_mode="updates",
):
    print(f"Receiving new event of type: {chunk.event}...")
    print(chunk.data)
    print("\n\n")

Install the LangGraph JS SDK:
npm install @langchain/langgraph-sdk

Send a message to the assistant (threadless run):

const { Client } = await import("@langchain/langgraph-sdk");

const client = new Client({ apiUrl: "your-deployment-url", apiKey: "your-langsmith-api-key" });

const streamResponse = client.runs.stream(
    null, // Threadless run
    "agent", // Assistant ID
    {
        input: {
            "messages": [
                { "role": "user", "content": "Say hello."}
            ]
        },
        streamMode: "messages",
    }
);

for await (const chunk of streamResponse) {
    console.log(`Receiving new event of type: ${chunk.event}...`);
    console.log(JSON.stringify(chunk.data));
    console.log("\n\n");
}

curl -s --request POST \
    --url <DEPLOYMENT_URL>/runs/stream \
    --header 'Content-Type: application/json' \
    --header "X-Api-Key: <LANGSMITH API KEY>" \
    --data "{
        \"assistant_id\": \"agent\",
        \"input\": {
            \"messages\": [
                {
                    \"role\": \"human\",
                    \"content\": \"Say hello.\"
                }
            ]
        },
        \"stream_mode\": \"updates\"
    }"

Next steps

Assistants

Deploy the same graph with different models, prompts, or tools per assistant.

Threads

Persist state across multiple runs so your agent remembers context between interactions.

Runs

Kick off background runs for long-running jobs and stream results back to your client.

Connect these docs to your agent of choice via MCP for real-time answers.

Edit this page on GitHub or file an issue.

Deployment components

Develop & test

Frameworks and platforms

Reference

Prerequisites

Get started with your coding agent

1. Create a deployable app

2. Set your API key

3. Deploy

4. Test in Studio

5. Test the API

Next steps

Assistants

Threads

Runs