Intel's new Gaudi 3 accelerators massively undercut Nvidia GPUs as AI race heats up

Intel's Gaudi 3 AI accelerators offer similar performance at half the cost of Nvidia's GPUs, challenging Nvidia's market dominance.

: Intel introduced its Gaudi 3 AI accelerators, dramatically undercutting Nvidia’s GPUs by up to 50%. The Gaudi 3 accelerators provide comparable performance in AI training and inference workloads. Despite the competitive pricing, Intel faces challenges in convincing enterprises to shift from Nvidia's CUDA platform.

Intel has launched its new Gaudi 3 AI accelerators, which are priced significantly lower than Nvidia’s H100 GPUs, with the flagship Gaudi 3 costing $15,000 per unit, nearly half the price of Nvidia’s offering. The Gaudi 2, a less powerful model, is also priced favorably at $65,000 for an 8-chip kit, which is just one-third of the cost of comparable Nvidia setups.

Intel claims that the Gaudi 3 offers similar or better performance than Nvidia's H100, with benchmarks showing up to 40% faster training times in large clusters and a 2x speed advantage in AI inference tasks. The company highlights that the Gaudi 3 leverages open standards like Ethernet for easier deployment, though it lacks the extensive software ecosystem of Nvidia’s CUDA platform, presenting a potential barrier to adoption.

To enhance its market penetration, Intel has partnered with major server vendors including Asus, Foxconn, and Dell. However, Nvidia still holds a dominant position with a 73% share of the data center processor market and an 88% share of the consumer GPU market. Intel’s lower pricing strategy might help it gain traction, but convincing enterprises to refactor their code away from CUDA will be a significant challenge.