Nous Research partners with Microsoft to bring Hermes Agent to the enterprise

Open original

Nous Research's open-source Hermes Agent is now available inside Microsoft Teams, the Microsoft Store, and Azure, aimed at companies that want a self-improving agent for every employee. The pitch is that successful workflows get saved as reusable skills the whole team can use.

  • Hermes Enterprise runs in Microsoft Teams, Hermes Desktop for Windows (via the Microsoft Store), and Azure Container Apps Sandboxes, where agent tasks run in isolated microVMs that start in under a second.
  • Hermes turns a successful approach to recurring work into a reusable skill, and a Collective Wisdom feature lets employees review and publish those skills for their team.
  • Companies with strict data residency or procurement needs can run Hermes in their own Azure tenant.

Nous Research, the team behind the open-source Hermes Agent, has launched Hermes Enterprise, bringing the agent into three Microsoft surfaces: Hermes Desktop for Windows in the Microsoft Store, Microsoft Teams, and execution on Azure Container Apps Sandboxes. In Teams, employees can hand Hermes work like financial analysis, drafting documents, or triaging an inbox, then review the finished output in the same conversation. The sandboxes run each workload in an isolated microVM that starts in under a second and releases compute when idle. For companies with strict data residency or procurement rules, the product can also run inside the customer's own Azure tenant under the same pricing model. The announcement leans on the fact that Hermes already ranks first among public open agents on OpenRouter, handling close to 2 trillion tokens a day. The more distinctive feature is what the company calls Collective Wisdom. When Hermes handles recurring work, it turns an approach that worked into a reusable skill, which is a written definition paired with code for the stable steps. Those skills run the same way each time, so routine workflows need less inference and get more reliable. Employees can ask Hermes which of their skills are worth sharing, then publish them to the team after review, so a new hire can start from the best method anyone has found. The pitch is that this tacit, institutional know-how stays a private asset of the company rather than becoming training data for an AI vendor. Whether that holds up in practice depends on the contract terms and the review process, neither of which the release spells out in detail. Microsoft's venture arm, M12, is the backer here, and its CVP Michelle Gonzalez framed the deal as a simple path to self-improving agents across teams. Nous will present at Microsoft Ignite in San Francisco on November 18, in session BRK448, on running isolated agent loops in production at scale. Prospective customers can find details at portal.nousresearch.com/business. Read it for the Teams and sandbox architecture if you're evaluating agent deployment, and treat the usage figures and the "gets better the more you use it" claim as vendor framing rather than independently verified results.

Ask about this article