Olares OS Integrates NVIDIA NemoClaw for Enhanced AI Functionality in Personal Hardware

At the recent GOSIM 2026 conference in Paris, the Olares team unveiled an exciting development: NVIDIA's NemoClaw is now capable of running natively on the Olares OS. This integration marks a significant milestone, allowing permissioned, self-evolving AI agents to operate seamlessly across various Olares-supported devices, including the newly launched Olares One desktop and NVIDIA's DGX Spark.

Olares OS is an open-source personal cloud operating system designed to transform local hardware into a secure alternative to public cloud services. It caters to a diverse range of users, including home lab enthusiasts, content creators, design studios, and teams focused on self-hosting, with a primary emphasis on maintaining control over data, models, and agent activities. The operating system, which is rooted in a lightweight Kubernetes framework, offers several key features, such as sandboxed application delivery, efficient GPU resource management, single sign-on capabilities, and access to over 200 curated one-click applications.

The introduction of NemoClaw builds upon this solid foundation, incorporating NVIDIA's OpenShell secure runtime specifically for autonomous agents. This means users can assign restricted permissions to agents concerning their data, network, and tools, providing a tailored approach versus broad trust models typical in cloud environments.

Olares One: The Engine for Local AI Workflows

The Olares One desktop is meticulously designed to facilitate comprehensive local AI workflows. This state-of-the-art desktop is equipped with an NVIDIA GeForce RTX 5090 Mobile GPU and 24 GB of GDDR7 memory. Olares OS enhances support across NVIDIA accelerated computing platforms, enabling it to run on various systems, including NVIDIA GeForce RTX graphics cards, NVIDIA RTX PRO workstations, and other advanced NVIDIA Blackwell devices like DGX B200 and DGX GB200. This versatility ensures that the entire agent stack can be operated locally on the user’s hardware model, encompassing model inference, retrieval, embeddings, and the management of NemoClaw-sandboxed agents—all while guaranteeing that no data or commands are sent off the device unless explicitly permitted by the user’s policies.

The Benefits of Kubernetes-Native Design

A key advantage of integrating NemoClaw with Olares OS lies in the Kubernetes-native architecture shared by both projects. This compatibility allows the Olares team to capitalize on a well-established set of components across the two systems. Consequently, this results in a reduced surface area for maintenance, quicker updates for upstream NemoClaw releases, and a cohesive operational experience for users, whether agents are functioning on an Olares One system in a studio, a DGX Spark on a developer's desk, or a Grace Blackwell machine in a dedicated rack.

Dr. Michael Yuan, Olares' VP of Research, articulated the significant value this integration brings to users: "Our users came to Olares because they refuse to hand their data, their models, or their agents to someone else's cloud. Pairing the OpenShell permissioning and sandboxing model with Olares OS across every NVIDIA GPU we support means the entire pipeline—from inference to autonomous agent—can finally run on hardware the user owns, and within boundaries they have set."

NVIDIA NemoClaw serves as an open-source reference stack derived from the OpenClaw project, aimed at simplifying the secure deployment of autonomous assistants. It offers a guided onboarding process, fortified blueprints, and implements OpenShell for layered process and network isolation, facilitating efficient, secure inference across both local and cloud-based models. This gives developers the capability to deploy always-on, self-evolving agents that adhere to the privacy and security policies defined by the user.

As of now, NemoClaw on Olares OS is available for immediate use, with source code, documentation, and installation instructions accessible at Olares GitHub and NVIDIA GitHub.

About Olares

Olares stands as a pioneer in the open-source personal cloud operating system marketplace, providing a private alternative to public cloud services for individuals, teams, and organizations. Supported by Olares ID and the LarePass client, Olares OS cultivates a growing ecosystem of self-hosted productivity, creative, and AI applications. This includes the recently launched Olares One desktop computer, which began shipping in April 2026 as a result of a successful Kickstarter campaign. To learn more about Olares and its innovative offerings, visit Olares Website.

In summary, the new compatibility between Olares OS and NVIDIA NemoClaw opens up exciting possibilities for AI development, significantly enhancing user autonomy in computing environments.