Broadcom has started shipping its newest networking chip, the Tomahawk 6, designed to dramatically enhance artificial intelligence (AI) performance by doubling the speed of its predecessor. Announced on June 3, the Tomahawk 6 not only boosts data throughput but also improves traffic control features, making it far more efficient for large-scale networking tasks, according to Ram Velaga, Broadcom’s Senior Vice President.
This upgrade significantly reduces the number of network switches needed to support AI infrastructure, helping lower complexity and potentially cost. As AI applications become more advanced, the supporting data centers require robust and highly scalable networking solutions. Broadcom’s Tomahawk chips have grown increasingly vital in this ecosystem, as they help interconnect tens of thousands of processing units required for AI workloads.
AI data centers, often composed of over 100,000 GPUs (graphics processing units), depend on advanced networking chips to function efficiently. The Tomahawk 6 is specifically engineered for this scale, enabling high-bandwidth, low-latency connectivity across vast GPU clusters. This makes it a crucial component in building modern AI supercomputers and hyperscale data centers.
“Tomahawk 6 is not just an upgrade – it’s a breakthrough. It marks a turning point in AI infrastructure design, combining the highest bandwidth, power efficiency, and adaptive routing features for scale-up and scale-out networks into one platform. Demand from customers and partners has been unprecedented. Tomahawk 6 is poised to make a rapid and dramatic impact on the deployment of large AI clusters,” said Ram Velaga, SVP and GM, Core Switching Group, Broadcom.
Broadcom’s move comes at a time when demand for AI-capable infrastructure is surging, with companies investing heavily in new chip designs and system architectures to support generative AI, machine learning, and other compute-intensive applications. With the launch of Tomahawk 6, Broadcom positions itself as a leading enabler of next-gen AI innovation by delivering essential hardware that facilitates faster, more efficient data transfer at scale.
This strategic release underscores the company’s commitment to advancing AI infrastructure and cementing its role in the high-performance networking chip market.
