Hermes Agent Maker Nous Research Nears $75M Round at $1.5B Valuation
Nous Research, the open-source lab behind the Hermes agent family, is finalizing a $75 million round led by Robot Ventures that would roughly 1.5x its valuation from last year's Series A.
Nous Research, the open-source AI lab known for its Hermes family of agent models, is finalizing a new funding round of at least $75 million at a $1.5 billion valuation, according to reports from TechCrunch and The Block published July 13. The round is led by Robot Ventures, with participation from Union Square Ventures and other existing backers.
From open-weights lab to $1.5B agent company
Founded in 2023 by Jeffrey Quesnelle, Karan Malhotra, Ryan Teknium and Shivani Mitra, Nous Research built its reputation on open-weight fine-tunes of frontier base models under the Hermes name, alongside Psyche, a decentralized training network that distributes model training across underutilized hardware rather than a single centralized GPU cluster. The company has previously raised roughly $70 million from investors including Paradigm, Robot Ventures, North Island Ventures, OSS Capital and Balaji Srinivasan, and was valued at around $1 billion in its Series A. The new round would mark a roughly 50% step-up in valuation in about a year.
Hermes as an enterprise agent
Hermes has evolved from a fine-tuning project into a standalone agent runtime that Nous Research and its investors are pitching as production-ready for enterprise use, including customer-service escalation and operational workflows. The most recent release in the line, Hermes 4.3, was the first Hermes model trained on the Psyche decentralized network rather than a conventional centralized cluster, with a post-training corpus that expanded roughly 50x in token count compared with the prior Hermes 4 release, according to the model's release notes.
Why it matters
The round underscores that investors continue to reward open-weight, decentralized alternatives to the closed frontier labs, even as the broader agent market consolidates around a handful of large platform players. Nous Research's pitch — an open, community-auditable agent stack paired with a training network that doesn't depend on hyperscaler GPU allocations — gives it a distinct lane from OpenAI, Anthropic and Google DeepMind's enterprise agent offerings, and the fresh capital suggests backers see room for that lane to scale into real enterprise deployments rather than remain a research curiosity.
Sources
AI-assisted reporting, overseen by the AgentsAI team. Spotted an error? Let us know.
More agents news
Nous Research Raises $90M Series B to Take Open-Source Hermes Agent to Enterprises
Nous Research confirmed a Series B to build Hermes for Businesses, an enterprise edition of its open-source Hermes Agent; the Wall Street Journal put the round at $90 million and a $1.5 billion valuation.
Oracle Launches Fusion Claw, a Governed Runtime for Autonomous Agents Inside Its ERP Apps
Oracle announced 25 Fusion Claw applications on September 29, built on an execution runtime with auditable 'outcome receipts' and role-based controls; availability is planned for Q4 2026 with consumption-based pricing.
OpenAI's Dots Bring Always-On Agents to ChatGPT Pro and Business Premium
Unveiled at DevDay 2026, dots are persistent GPT-6 Astra agents with their own cloud computer that work toward user goals around the clock. They are rolling out to ChatGPT Pro and Business Premium, with the EU, Switzerland and UK excluded for Pro.
Apple Will Tighten macOS Full Disk Access, Citing Risks From AI Agents
Apple says it will add controls so Mac apps can only get Full Disk Access through 'very explicit user action', warning that the risks grow as AI agents become more autonomous. No timeline was given.