Skip to main content
Hexel Studio provides the infrastructure to deploy and run AI agents in production. Register a Docker image, deploy it, and get a permanent endpoint.

Quickstart

Deploy your first agent in 5 minutes.

Compute Overview

Learn about sandboxes, agents, and instances.

SDKs and CLI

Python, TypeScript, and command-line tools.

API Reference

REST API documentation.

​
How it works

Hexel Compute has two modes: Agent Hosting. Deploy a Docker image and get a permanent URL. Your agent stays in memory with millisecond wake times. Authentication is handled by the platform. Sandboxes. Allocate an isolated execution environment instantly. Run Python code, shell commands, and file operations. Release when done.

​
Example

from hexel import Hexel

client = Hexel(api_key="your_api_key")

# Run code in a sandbox
sandbox = client.compute.sandbox.create(tier="standard")
client.compute.sandbox.execute(sandbox["vm_id"], code="print('Hello')", language="python")
client.compute.sandbox.release(sandbox["vm_id"])

# Deploy an agent
agent = client.compute.agent.register(name="my-agent", image="ghcr.io/org/agent:v1")
instance = client.compute.instance.deploy(agent["id"], env={"MODEL": "gpt-4"})
# Live at: https://zeal-isle-a620.compute.hexelstudio.com

​
Getting started

1

Create an account

Sign up at console.hexelstudio.com.
2

Create an API key

Go to IAM, create a Service User, and generate an API key.
3

Install

Choose your tool: pip install hexel, npm install hexel-sdk, or curl -fsSL https://hexelstudio.com/install.sh | bash.
4

Follow the quickstart

See the quickstart guide for a complete walkthrough.