Hexel Studio Overview - Hexel Studio

Hexel Studio is the infrastructure layer for AI agents in production. You ship agent code as a Docker image and Hexel gives it a permanent endpoint, isolated compute, searchable knowledge, persistent memory, connections to external tools like Gmail and GitHub, and multi-agent orchestration. Instead of stitching together queues, secret stores, vector databases, and container orchestrators yourself, you get a single API that covers the full lifecycle of an agent: from a prototype running code in a sandbox to a fleet of agents triaging support tickets or processing invoices.

Quickstart

Run code and deploy your first agent in under 10 minutes.

Authentication

Create an API key and authenticate the SDKs, CLI, or REST API.

Core Concepts

Agents, fleets, tasks, tools, knowledge, and organizations.

API Reference

The full REST API with an interactive playground.

What you can build

Compute

Deploy agents with permanent endpoints and run isolated sandboxes for code execution.

Data Platform

Give agents knowledge stores, persistent memory, and connectors to your data.

Orchestration

Submit tasks to agent fleets that plan, execute, and stream results as a workflow.

Tool Gateway

Connect agents to Gmail, Slack, GitHub, Jira, and more with managed OAuth.

Deploy in two steps

Python
CLI
curl

from hexel import Hexel

client = Hexel(api_key="YOUR_API_KEY")

agent = client.compute.agent.register(
    name="support-agent",
    image="ghcr.io/your-org/support-agent:v1",
    capabilities=["chat", "streaming"],
)
print(agent["id"])

hexel compute agent register \
  --name support-agent \
  --image ghcr.io/your-org/support-agent:v1 \
  --capabilities "chat,streaming"

hexel compute instance deploy YOUR_AGENT_ID

curl -X POST https://api.hexelstudio.com/compute/v1/agents/YOUR_AGENT_ID/instances \
  -H "Authorization: Bearer $TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"env": {"MODEL": "gpt-4"}}'