AMD's Helios AI Server Enters Full Production with Q3 Shipment Target

AMD CEO Lisa Su announced that the company's second-generation AI server, Helios, has entered full production and is scheduled to ship by the end of Q3. The server will feature the new Instinct MI455X AI accelerators and next-generation Venice CPUs, both manufactured by TSMC. This release intensifies AMD's competition with NVIDIA in the lucrative data center market, especially with major players like OpenAI planning to deploy Helios later this year. It represents a significant push by AMD to capture a larger share of the rapidly growing AI hardware market, which is projected to reach trillions of dollars by 2030. The Helios server is designed as an open, modular rack-scale system packing 72 MI455X GPUs, which utilize TSMC's 2nm (N2) process and feature 432GB of HBM4 memory. The Venice CPUs powering the system are based on the Zen 6 architecture and can scale up to 256 cores.