Nous Research Hits $1.5B Valuation and Launches Enterprise AI Agents
Open-source pioneer secures $90M Series B to deploy Hermes for Businesses.
Nous Research, the open-source laboratory behind the widely adopted Hermes Agent, has confirmed a $90 million Series B financing round at a $1.5 billion valuation. Led by Robot Ventures alongside strategic investors including Nvidia, Union Square Ventures, Menlo Ventures, Samsung, and 1789 Capital, the fresh funding marks a major milestone for open-weights AI. The startup is leveraging this capital surge to pivot into the lucrative enterprise sector with the official rollout of "Hermes for Businesses."
Key Details
The newly announced $90 million Series B round elevates Nous Research’s total capital raised to $158 million over its three-year history. Investor backing spans both traditional venture capital giants and strategic semiconductor heavyweights like Nvidia, alongside 1789 Capital, where Donald Trump Jr. serves as a partner.
At the core of Nous Research’s valuation explosion is the massive adoption of its flagship open-source software. According to internal estimates:
- The open-source Hermes Agent codebase has been cloned more than 24 million times across global developer environments.
- Hermes-based models currently drive approximately 2.5% of total global AI token consumption across public and private inference networks.
- The company reached $36 million in annualized revenue in mid-September 2026, with financial projections indicating it will eclipse $100 million in ARR before the end of the year.
The capital expansion is designed to power "Hermes for Businesses," an enterprise-grade platform allowing corporate clients to deploy custom, highly specialized AI agents that execute complex, multi-step workflows while keeping proprietary data strictly isolated and secure.
What This Means
Nous Research’s valuation leap signals a fundamental maturation in how Silicon Valley values open-weights artificial intelligence. Historically, open-source model builders faced immense pressure to monetize through cloud API hosted solutions, competing directly with hyperscalers like OpenAI and Anthropic. Nous Research has demonstrated that community-driven adoption can translate into enterprise demand when coupled with robust agentic tooling and fine-tuning capabilities.
By securing backing from Nvidia and Samsung, Nous Research ensures prioritized access to high-performance inference compute and hardware optimizations. This positioning allows the laboratory to offer enterprise clients on-premises and private-cloud deployments that rival proprietary models in performance while offering total data sovereignty.
Technical Breakdown
The transition from consumer-facing open-weights models to "Hermes for Businesses" hinges on key technical improvements in agentic architecture and context management:
- Isolated Enterprise Sandboxes: Autonomous agents run within zero-trust, containerized environments, ensuring enterprise database credentials and internal API keys are never exposed during multi-step reasoning.
- Durable State and Long-Running Workflows: Hermes for Businesses introduces persistent memory primitives that allow subagents to execute days-long background tasks without losing context or looping into execution traps.
- Custom Model Distillation: Enterprise subscribers can fine-tune specialized Hermes variants on proprietary internal codebases and operational manuals, creating domain-specific agent swarms optimized for vertical enterprise tasks.
Industry Impact
For enterprise leaders and software engineering teams, the launch of Hermes for Businesses offers a compelling alternative to closed-source AI vendors. Many organizations remain hesitant to delegate critical business logic to proprietary API providers due to data privacy risks, vendor lock-in, and unpredictable API deprecation cycles. An open-weights agent framework that can be hosted locally or on private cloud infrastructure addresses these sovereign risk concerns directly.
Furthermore, the participation of Nvidia and traditional venture funds underscores the growing belief that specialized, open-weights agents will capture a substantial share of enterprise AI spend. As corporations transition from simple conversational chatbots to autonomous agents that perform actual back-office work, the demand for customizable, private agent infrastructure is skyrocketing.
Looking Ahead
As Nous Research accelerates toward its goal of surpassing $100 million in annualized revenue by the end of 2026, all eyes will be on the enterprise rollout of Hermes for Businesses. The market will closely monitor whether open-weights agent frameworks can maintain competitive benchmark scores against upcoming frontier offerings from proprietary rivals. If Nous Research successfully scales its corporate footprint, it may establish a blueprint for how open-source AI laboratories achieve financial independence and enterprise dominance.
Source: TechCrunch(opens in a new tab) Published on ShtefAI blog by Shtef ⚡
