What started as simple chat interfaces has quickly become persistent, autonomous systems that can code, manage files, connect to APIs, learn, and operate across multiple environments.

That’s why we’re excited to share that Pinata Agents now supports Hermes.

Pinata Agents make it easy to deploy hosted agents from the start with workspaces, isolated containers, skills, and deployable templates from our industry partners. Originally launching with OpenClaw, now you can also launch and manage Hermes-powered agents directly on Pinata’s trusted infrastructure.

What is Hermes?

Hermes Agent is an open-source AI agent that self-improves. Built by Nous Research, it’s currently the only agent with a built-in learning loop, so it’s designed to function more like a continuously evolving operating environment for AI agents.

With Hermes supported inside Pinata Agents, you can get hosted infrastructure for deploying these agents without needing to manually manage containers, runtimes, or orchestration layers themselves.

What Pinata Agents Does

Pinata Agents provides managed infrastructure for deploying AI agents. Deployed agents can:

• Run code
• Search the web
• Manage files
• Connect to external services
• Store persistent state
• Use templates and reusable skills
• Expose routes and integrations
• Operate continuously over time

Instead of spending time configuring environments, provisioning infrastructure, and wiring together tooling, you can focus on building the behavior and workflows to make your agents useful.

Whether you want to launch a coding assistant, research system, operations workflow, autonomous monitoring agent, or long-running AI environment, Pinata Agents gives you the infrastructure layer to make it happen.

And now, you can do it with Hermes, too.

Start building at https://agents.pinata.cloud/ and check out the templates at https://agents.pinata.cloud/landing/marketplace.