Nvidia Plans Open-Source AI Agent Platform ‘NemoClaw’ for Enterprise Use

Nvidia is reportedly preparing to introduce a new open-source platform for artificial intelligence agents called “NemoClaw,” signaling the chipmaker’s increasing push into enterprise AI software. According to a report from Wired, the company has already begun presenting the platform to several major enterprise software firms as it looks to build a broad ecosystem around the technology.

The initiative reflects a growing shift in the artificial intelligence industry toward so-called AI agents—systems designed not just to generate text but to plan, reason and execute multi-step tasks independently. These agents can automate complex workflows for businesses, potentially transforming how employees interact with software tools.

Sources familiar with the matter said Nvidia has been discussing possible partnerships with companies such as Salesforce, Cisco, Google, Adobe, and CrowdStrike. While no formal agreements have been confirmed, the goal is reportedly to encourage these companies to integrate the platform into their enterprise products.

Because NemoClaw is expected to be open source, partners may be able to use the system without paying licensing fees. Instead, early participants could gain access in exchange for contributing code, tools or integrations to the project. This approach could help Nvidia rapidly expand the platform’s capabilities while building a community around its development.

The platform is designed to allow organizations to deploy AI agents capable of completing tasks on behalf of employees, such as analyzing data, coordinating workflows or interacting with other software services. Security and privacy features are also expected to be included, addressing concerns that many companies have about deploying autonomous AI tools inside corporate environments.

Importantly, the platform will reportedly be hardware-agnostic, meaning companies can use it regardless of whether their systems run on Nvidia chips. That flexibility could make NemoClaw more appealing to enterprises that operate mixed technology environments.

Nvidia’s interest in AI agents builds on its broader investment in artificial intelligence software. In recent months, the company has introduced models such as Nemotron and Cosmos, designed specifically to support agent-based AI systems. It has also expanded its Nvidia NeMo platform, which provides tools for managing the entire AI lifecycle—from training models to monitoring their performance in real-world applications.

The push toward agent technology has been fueled by the rapid rise of open-source projects known informally as “claws.” These tools allow AI agents to run locally on a user’s machine and carry out sequences of tasks automatically.

One of the most prominent examples is OpenClaw, a project that quickly gained attention earlier this year for its ability to automate complex workflows. The project’s success ultimately led OpenAI to acquire it and bring its creator onto the team.

Nvidia CEO Jensen Huang has previously praised OpenClaw, describing it as one of the most significant software releases in recent years. However, cybersecurity experts have also warned that early AI agents may introduce new risks, particularly if they are given broad access to corporate systems and data.

The reported launch of NemoClaw comes just ahead of Nvidia’s annual developer conference in San Jose, where the company is expected to unveil new updates to both its hardware and AI software platforms. Industry observers say the event could provide more details about Nvidia’s long-term strategy for enterprise AI and autonomous software agents.