Persistent memory
Completed work becomes a reusable skill. The tenth time a task comes round it is a replay, not a fresh chain of reasoning: faster, cheaper and more consistent.
An agent that
keeps what it learns.
An MIT-licensed agent from Nous Research that remembers how it solved something and does it faster the next time. We deploy and operate it for you.
Discuss your setup 01Deploy02Connect03Teach04Maintain
Deployed, scoped and looked after.
Where your team already is
Reachable on Telegram, Discord, Slack, WhatsApp, Signal, email and the command line. No new app for anyone to remember to open.
Model agnostic
Runs on Claude, GPT, Gemini, DeepSeek, Qwen or any OpenAI-compatible endpoint, so a change of provider is a config change rather than a rebuild.
Infrastructure you own
MIT licensing and self-hosting mean no per-seat lock-in. The agent, its memory and its credentials sit on infrastructure you own.
Taught your workflows
We seed the first skills against your real workflows, so the memory is useful from week one rather than month six.
Kept healthy
Upgrades, model changes, memory review and monitoring, handled on an ongoing basis.
Looking for a tool gateway?
OpenClaw connects your models to tools and messaging surfaces.
Explore OpenClawBefore we get started.
Is Hermes Agent a Nexum product?
No. Hermes Agent is an open-source, MIT-licensed project from Nous Research, released in February 2026. Nexum’s work is deploying, configuring and operating it on your behalf.
Where does it run?
Self-hosted on your servers or cloud tenancy, with memory storage configured to your retention rules.
What can it reach?
The messaging channels your team actually uses, plus the tools and accounts the agent needs, and nothing beyond them.
How is it different from OpenClaw?
Hermes is built around persistent memory and reusable skills. OpenClaw is a gateway that connects models to tools and messaging surfaces. We can help you choose, or run both.