Cornelis Networks raises $205M for high-performance AI fabric

The money is earmarked for scale-up networking, the next-generation scale-up and scale-out roadmap, and CN6000 manufacturing and customer deployments. The CN6000 hardware is already sampling with customers. The round was announced alongside the Active Compute Fabric, an open architecture spanning scale-up and scale-out networking with programmable compute built into the fabric. It also came with a Qualcomm Technologies collaboration on validation toward future rack-scale AI data center designs. Existing backers including Foresight Group, Intel Capital, KittyHawk Ventures, Ridgeline Partners and SQN Venture Partners also took part. A $112M convertible note closed in December 2024. The company competes with established providers such as Cisco Systems and Arista Networks, and its pitch targets Nvidia's proprietary interconnects. An IAG partner put the open-standard scale-up and scale-out opportunity at more than $55 billion by 2030. For teams building AI companies, the practical point is supply and cost. Networking increasingly decides how much of a GPU cluster's capacity gets used, and a funded open-standards alternative could give buyers leverage against single-vendor stacks. The caveat is maturity: the newest silicon is still at the sampling stage, so any near-term deployments depend on how well the CN6000 performs in customer trials.