Prime Intellect Raises $130M in Series A Funding to Empower Enterprises in Creating Custom AI Agents

Prime Intellect, a startup focused on providing computing power and customized software tools to help companies create AI agents, has successfully completed a $130 million Series A funding round, achieving a valuation of $1 billion.

This significant investment was led by Radical Ventures, with participation from Nvidia Ventures, Intel Capital, Dell Technologies Capital, Iconiq, and several prominent angel investors, such as Aravind Srinivas (Perplexity), Aaron Levie (Box), Winston Weinberg (Harvey), Jeff Wang (Cognition), and Brendan Foody (Mercor).

Founded in 2024, Prime Intellect aims to enable organizations to independently train their own autonomous systems without relying on high-end AI labs. This once ambitious goal is now achievable due to breakthroughs in reinforcement learning techniques that reward successful task completion and penalize errors, allowing companies to become their own “AI labs” by refining models for unique business needs.

Despite the potential to operate outside closed AI labs, the complex underlying infrastructure is a hurdle for many organizations, which frequently lack the expertise needed to integrate these components into a production-ready system.

This is where Prime Intellect becomes essential.

The startup has developed what it calls a comprehensive solution for AI agent development, which includes computing resources, a reinforcement learning framework, and evaluation tools.

Prime Intellect’s platform functions as a marketplace, providing modular access so clients can choose the specific tools they need without a commitment to an all-inclusive package.

“They’ve assembled this in a way that maintains their position at the cutting edge while remaining budget-friendly,” remarked David Katz, a partner at Radical Ventures. He further noted that while others might offer piecemeal solutions, Prime Intellect distinguishes itself by offering the complete capabilities of a top-tier AI lab as a “one-stop shop” for development.

The startup’s business model has attracted clients like Ramp, Zapier, and Flapping Airplanes, who subscribe to a hosted version of its tools. This rapid adoption has elevated the company to an annual revenue run rate of $100 million.

This growth is driven by significant results. For example, Ramp used Prime Intellect to create an agent capable of deriving insights from spreadsheets. “The results exceeded those of leading models in accuracy while operating at quicker speeds and a fraction of the cost,” stated Ramp’s co-founder and co-CEO Karim Atiyeh.

A crucial factor in Prime Intellect’s expansion is the growing awareness among companies regarding the risks tied to reliance on frontier labs.

Organizations are becoming more reluctant to share proprietary data with firms like OpenAI and Anthropic, fearing loss of control over their information. Apprehension also exists about depending on models that could suddenly become inactive, as illustrated by Anthropic’s Fable incident last month.

“How can I trust that I’m not working with a company that might replace me or generalize my work?” Katz commented. “All these considerations are pushing businesses to ponder, ‘How can I secure ownership of my enterprise intelligence while mitigating these risks?’”

Vincent Weisser, co-founder and CEO of Prime Intellect, believes that enterprises are eager to transition away from closed-source frontier models, and his company provides the infrastructure necessary to support that shift.

“It shouldn’t be restricted to a few experts in a glass tower in San Francisco who can train AI models,” he shared with TechCrunch. “This capability should be accessible to every enterprise, every nation state.”