Agent Studio is currently in alpha. Not all features mentioned are fully available yet.

Next-Gen Agent Platform

Build, Host & Test Your LangGraph Workflows

Agent Studio is the complete platform for LangGraph agent developers to upload, run, and test their graph workflows with ease.

langgraph_workflow.py
from langchain.langgraph import Graph
from langchain.agents import AgentExecutor
from langchain.chat_models import ChatOpenAI

# Define your agent workflow
def create_workflow():
    llm = ChatOpenAI(temperature=0)
    tools = [...]
    agent = AgentExecutor.from_llm_and_tools(
        llm=llm,
        tools=tools
    )

    workflow = Graph()
    workflow.add_node("agent", agent)
    # Add more nodes and edges...

    return workflow
Features

Everything You Need for LangGraph Development

Agent Studio provides a complete platform for developing, hosting, and testing your LangGraph workflows.

Full Hosting

Deploy your LangGraph workflows with zero infrastructure management.

API Keys

Generate API keys to easily integrate your workflows into any application.

Compatible API

Same outputs as LangGraph's server API for seamless integration.

Instant Testing

Test your workflows directly in the platform with real-time feedback.

Version Control

Track changes and manage different versions of your workflows.

SDK Support

Full support for @langchain/langgraph-sdk and @langchain/core.

How It Works

Simple Workflow for Complex Agents

Agent Studio makes it easy to deploy and manage your LangGraph workflows.

1

Upload

Upload your LangGraph workflow to Agent Studio using our simple interface or CLI.

2

Configure

Set up your environment variables, API keys, and deployment settings.

3

Deploy

Deploy your workflow with a single click and get an API endpoint instantly.

Integration Example

integration.js
import { LangGraphClient } from "@langchain/langgraph-sdk";

// Connect to your Agent Studio workflow
const client = new LangGraphClient({
apiUrl: "https://api.agentstudio.dev/v1",
apiKey: "your_api_key"
});

// Use the workflow in your application
const response = await client.invokeGraph({
graphName: "my-workflow",
inputs: { query: "What is the weather today?" }
});
Pricing

Simple, Transparent Pricing

Choose the plan that's right for your development needs.

Free

Perfect for testing and small projects

$0/month
  • 3 workflows
  • 1,000 API calls/month
  • Community support
Popular

Pro

For professional developers and teams

$49/month
  • Unlimited workflows
  • 50,000 API calls/month
  • Priority support
  • Advanced analytics

Enterprise

For large organizations with custom needs

Custom/month
  • Unlimited everything
  • Dedicated support
  • Custom integrations
  • SLA guarantees
FAQ

Frequently Asked Questions

Answers to common questions about Agent Studio.

What is LangGraph?

LangGraph is a framework for building stateful, multi-actor applications with LLMs. It extends LangChain with a graph-based architecture for creating complex agent workflows.

How does Agent Studio work with LangGraph?

Agent Studio provides hosting and management for your LangGraph workflows. You can upload your workflows, and we'll handle the infrastructure, scaling, and API endpoints.

Can I use my existing LangGraph code?

Yes! Agent Studio is designed to be compatible with existing LangGraph code. You can use @langchain/langgraph-sdk and @langchain/core just as you would with LangGraph's server API.

Do you support custom tools and models?

Absolutely. Agent Studio supports all the tools and models that LangGraph supports, including custom tools and integrations with various LLM providers.

How do I get started?

Sign up for an account, upload your LangGraph workflow, and you'll get an API endpoint that you can use in your applications. It's that simple!

What about security and privacy?

We take security seriously. Your workflows are isolated, and we provide secure API keys for access. We also offer enterprise-grade security features for larger organizations.

Ready to Get Started?

Join the growing community of LangGraph developers using Agent Studio.