Prime Intellect Raises $130M to Scale AI Agent Training Platform

Prime Intellect secures major funding for AI training infrastructure

Prime Intellect has closed a $130 million Series A round to expand its platform that enables companies to train and deploy their own AI agents using reinforcement learning technology. The round was led by Radical Ventures, with participation from Nvidia Ventures, Intel Capital, Dell Technologies Capital, and existing investors, according to a company blog post published July 8.

The funding brings Prime Intellect's total capital raised to over $150 million. The company previously announced a $15 million raise in February 2025 that brought its total funding at that time to $20 million.

Prime Intellect provides customers with compute resources, large-scale reinforcement learning capabilities, environments, sandboxes, evaluations, and deployment tools needed to train, deploy, and continuously improve AI agents. The company currently serves 6,000 customers spanning AI startups, research labs, and enterprises, and has reached over $100 million in annualized revenue.

Why it matters

Reinforcement learning represents a fundamental shift in how companies can develop AI systems. Unlike traditional pre-training approaches that concentrate frontier AI development in a handful of large labs, RL enables individual companies to own their model optimization process. Organizations can now train models directly on their own products, optimize for specific workflows, and build agents that improve continuously in production environments. This democratization of advanced AI training could accelerate innovation across industries while allowing companies to maintain proprietary advantages in their AI implementations.

Scaling reinforcement learning infrastructure

The new capital will fund expansion of Prime Intellect's infrastructure to support larger compute clusters and larger RL training runs. The company plans to build out its stack for agentic training, inference, and continual learning capabilities.

Prime Intellect also intends to develop infrastructure for complex challenges including long-horizon agents, recursive language models, automation of AI research and science, and continual learning systems.

Intel Capital Principal Alexandra Farmer and Investment Director Assaf Araki described Prime Intellect as representing "the next generation of AI infrastructure" in a blog post. They noted that while reinforcement learning is emerging as a new method to generate data and train models for specific tasks, running RL on large language models proves far more complex than standard fine-tuning approaches.

"Going forward, we believe every AI builder will need reliable RL infrastructure to create competitive models and products, accelerating the demand for RL tooling," Farmer and Araki wrote.

The company emphasized in its announcement that pre-training has historically concentrated frontier AI development in a small number of labs, but reinforcement learning breaks that pattern open by allowing companies to control their own model optimization loops.

These details were first reported by PYMNTS.