Cornelis Raises $205M for AI Networking Tech Aimed at Nvidia’s GPU Stack Advantage

Cornelis, a company building networking technology to help AI chips communicate more effectively, announced that it raised $205 million. The round was led by IAG Capital Partners. The raise puts fresh attention on the infrastructure layer around AI chips, where performance depends not just on compute power but on how efficiently data moves between hardware.

Cornelis also announced Active Compute Fabric, a network technology aimed at reducing GPU time wasted while chips wait for data to arrive. The company’s approach is to let chips process and send information at the same time. For AI teams, the core takeaway is that networking fabric is becoming a key performance battleground, not just a back-end detail.

Cornelis spun off from Intel in 2020 and is competing with Nvidia through an open architecture model. That means customers can use a variety of GPU and accelerator hardware with Cornelis’ networking fabric. Nvidia chips can technically run on other networking fabrics, but they are optimized for Nvidia’s own software, which makes the full Nvidia GPU stack easier and more appealing for customers.

Cornelis has already started shipping its product and is working on a new generation expected later this year. The company is part of a broader wave of AI infrastructure startups trying to break apart Nvidia’s market dominance piece by piece. The immediate question is whether open networking options can give customers more flexibility without sacrificing performance.