Prime Intellect raises $130M Series A to help enterprises build their own AI agents at $1B valuation

Prime Intellect Secures $130M Series A to Democratize AI Agent Development

Prime Intellect has closed a $130 million Series A funding round at a $1 billion valuation, positioning itself as a critical player in helping enterprises build their own AI agents without depending on closed frontier labs 1

1

. The massive round was led by Radical Ventures, with participation from Nvidia Ventures, Intel Capital, Dell Technologies Capital, Iconiq, and notable angel investors including Perplexity's Aravind Srinivas, Box's Aaron Levie, and Harvey's Winston Weinberg 1

1

.

Founded in 2024, the startup has rapidly scaled to serve 6,000 customers and achieve an annualized revenue run rate of $100 million 2

2

. This growth reflects a fundamental shift in how organizations approach AI development, moving away from reliance on companies like OpenAI and Anthropic.

Full-Stack Platform Enables Organizations to Train AI Agents Independently

Prime Intellect has developed what CEO Vincent Weisser describes as a comprehensive infrastructure that allows companies to become their "own AI lab" 1

1

. The full-stack platform includes compute access, reinforcement learning frameworks, environments, sandboxes, evaluations, and deployment tools needed for AI agent development 2

2

.

The platform functions as a marketplace where customers can select specific tools without being locked into an all-or-nothing system. David Katz, a partner at Radical Ventures, noted that while others offer fragments, Prime Intellect is unique in providing the capabilities of a top-tier AI lab as a "one-stop shop" for development 1

1

.

Reinforcement Learning Breaks Open Concentrated AI Power

The rise of reinforcement learning techniques has made it possible for enterprises to build their own AI agents by refining open frontier models for specific business tasks 1

1

. These techniques iteratively reward successful task completion and penalize errors, allowing companies to optimize models for their unique workflows.

"Pre-training concentrated frontier AI in a handful of labs," Prime Intellect stated. "RL breaks that open: companies can now own their model optimization loop -- train directly on their own product, optimize for their specific workflows and build agents that improve continuously in production" 2

2

.

Intel Capital Principal Alexandra Farmer and Investment Director Assaf Araki emphasized that reinforcement learning is emerging as a new way to generate data and train models, though running RL on large language models is far more complex than standard fine-tuning 2

2

.

Tangible Results Drive Rapid Customer Adoption

The startup's approach has attracted prominent customers like Ramp, Zapier, and Flapping Airplanes, who pay for a hosted version of its tools 1

1

. Ramp used Prime Intellect to build an agent that finds answers inside spreadsheets. "The result beat the frontier models on accuracy while running at faster speeds and a fraction of the cost," said Ramp's co-founder and co-CEO Karim Atiyeh 1

1

.

Growing Concerns About Frontier Lab Dependence Fuel Demand

Companies increasingly hesitate to provide proprietary information to OpenAI and Anthropic due to risks of losing control over their data . Organizations also worry about depending on models that can be suddenly discontinued, as happened with Anthropic's Fable last month 1

1

.

"How do I know that I'm not working with a company that is going to try to replace me and generalize to what I'm doing," Katz explained. "All of these things are causing people to think, 'How do I own my own enterprise intelligence and not have these risks'" 1

1

.

Scaling AI Infrastructure for Agentic Systems and Continual Learning

With the new funding, Prime Intellect will scale its stack to include larger compute clusters, larger RL runs, and infrastructure for agentic training, inference, and continual learning 2

2

. The company will also build infrastructure for long-horizon agents, recursive language models, automation of AI research and science, and proprietary AI solutions 2

2

.

The Series A round brings Prime Intellect's total funding to over $150 million, following a $15 million raise in February 2025 2

2

. Weisser believes enterprises are looking to move away from closed-source frontier models: "It shouldn't just be a few nerds in a glass tower in San Francisco that have the capability to train AI models. It should be every enterprise, every nation state" 1

1

.