Nous Research Raises $90M to Bring Open-Source Hermes AI Agent to Businesses

Nous Research, the AI startup behind the open-source Hermes Agent, has raised $90 million in Series B funding to bring its AI agent technology to enterprise customers. The round was led by Robot Ventures, with participation from NVIDIA, M12, Samsung, Y Combinator, Union Square Ventures and Menlo Ventures, among other investors.

According to The Wall Street Journal, the financing values Nous Research at approximately $1.5 billion. The company has not independently confirmed that valuation. TechCrunch reported that the latest round brings its total funding to $158 million.

The company plans to use the new capital to develop Hermes for Businesses, an enterprise-focused version of its open-source AI agent designed to help organisations automate tasks while maintaining greater control over their models, data and institutional knowledge.

From open-source AI research to enterprise software

Nous Research was founded in 2023 by Jeffrey Quesnelle, Karan Malhotra, Ryan Teknium and Shivani Mitra.

The company initially gained attention through its work fine-tuning Meta's Llama models and publishing open AI research. It later developed Hermes Agent, releasing the software under the permissive MIT licence in February 2026.

Hermes is designed to do more than answer questions. According to its documentation, the agent can retain memory across sessions, create reusable skills, search the web, execute code and manage scheduled tasks.

Users can run the software themselves or pay for hosted plans. This open-source approach gives developers and organisations flexibility in how they deploy the agent.

Nous says Hermes has been cloned more than 24 million times. The Wall Street Journal reported more than 22 million downloads, while the company estimates that its software accounts for approximately 2.5% of global AI token usage. The latter figure is based on the company's internal estimates.

These numbers point to growing interest in the project, although downloads and code clones do not necessarily translate into active users or paying customers.

What is Hermes for Businesses?

The latest funding will support Hermes for Businesses, which is intended to bring the open-source agent into corporate environments.

Nous says the enterprise product will allow organisations to choose AI models based on cost and task requirements, keep their AI usage private and retain ownership of institutional knowledge.

The ability to select different models could help businesses balance performance, cost and specific workflow requirements rather than relying on a single AI provider.

Memory and reusable skills are also central to the product's approach. An agent that can retain useful context and repeat workflows may help employees automate recurring work instead of rebuilding instructions for every task.

Nous also plans to introduce a mobile app, although the company has not announced a release date.

The commercial opportunity lies in making Hermes easier to deploy and manage for organisations that need more control than a standard consumer AI assistant provides.

Why enterprise customers may pay for open-source AI

Open-source software creates an interesting business challenge: if companies can download and operate the software themselves, why would they pay its developer?

For Nous, the answer is enterprise infrastructure and control.

Businesses adopting autonomous AI agents must consider privacy, access permissions, security, governance and the handling of internal information. Agents may need access to company documents, software systems and credentials to complete tasks.

That creates operational risks alongside the productivity benefits.

Nous is positioning Hermes for Businesses around private deployment, model flexibility and control over institutional knowledge. These capabilities could appeal to companies that want to use AI agents without giving up control over how their systems operate.

However, the company will need to demonstrate that its enterprise offering delivers enough value to justify a paid service when the underlying software is freely available.

A growing race to build AI agents for companies

Nous Research is entering a competitive market that includes open-source projects and commercial AI platforms.

OpenClaw is another open-source agent project that has attracted attention. OpenAI hired its creator, Peter Steinberger, in February 2026, according to Bloomberg reporting cited in the supplied article.

NVIDIA has also developed an enterprise platform based on OpenClaw, according to The Next Web.

Meanwhile, OpenAI and Anthropic offer their own commercial AI agent products.

The competitive field also includes startups focused on specific enterprise workflows.

Dust raised $40 million in Series B funding in May 2026 for a workspace where employees and AI agents can share context. Bespoke Labs raised $40 million across seed and Series A rounds to develop simulated company environments in which agents can practise complex tasks.

Berlin-based telli raised $15 million in seed funding in July for AI agents that manage customer conversations across voice, WhatsApp, chat and email.

Stuut raised $52.5 million in Series B funding in October 2026 to develop AI agents for enterprise receivables and collections.

These companies approach enterprise AI from different directions, but they share a broader ambition: moving AI beyond isolated prompts toward systems that can complete tasks within real business workflows.

Investors see potential in enterprise AI agents

Robot Ventures led Nous Research's $90 million Series B. Other named backers include NVIDIA, Microsoft's M12 venture fund, Samsung, Y Combinator, Union Square Ventures and Menlo Ventures.

TechCrunch also reported that 1789 Capital participated in the round.

The financing follows Nous Research's $50 million Series A, led by Paradigm in April 2025, and an earlier $8 million seed round listed by Dealroom.

The latest investment gives the company additional resources to build enterprise features and expand its commercial offering.

The reported $1.5 billion valuation and approximately $36 million annualised revenue figure suggest significant investor expectations. However, the valuation was reported by The Wall Street Journal rather than confirmed by Nous, and the revenue figure should be treated as reported annualised revenue rather than audited annual results.

The enterprise AI agent market is expanding

Research firm Grand View Research estimates that the global AI agents market was worth $7.6 billion in 2025 and could reach $182.9 billion by 2033.

The firm also projects that the enterprise agentic AI market could grow from $5.3 billion in 2026 to $24.5 billion by 2030.

These are market forecasts, not guaranteed outcomes. Still, they reflect the growing commercial interest in AI systems that can execute multi-step tasks rather than simply generate text or respond to questions.

For Nous Research, the challenge is converting the adoption of its open-source software into a sustainable enterprise business.

Can Nous turn free software into a major business?

Hermes Agent has built visibility through its open-source distribution model. The new funding gives Nous Research an opportunity to develop a paid offering around that technology.

The company's strategy depends on persuading businesses that enterprise deployment, privacy, model choice and management capabilities are worth paying for, even when the underlying agent can be accessed for free.

It must also show that its agents can operate reliably and securely in corporate environments, where mistakes involving data, permissions or automated actions can carry significant consequences.

With $90 million in new funding and a reported valuation of $1.5 billion, Nous Research is making a substantial bet on enterprise demand for open-source AI agents.

Its next challenge is to prove that widespread developer adoption can translate into recurring revenue from business customers.