In a bold step toward advancing enterprise-level artificial intelligence, IBM has introduced its newest generation of mainframe technology—the IBM z17. Designed with AI at its core, the z17 integrates advanced processing capabilities to support over 250 AI-driven applications, including generative models and intelligent agents.
This next-generation mainframe is equipped with the IBM Telum II processor and features end-to-end encryption, giving organizations both high performance and robust security. According to IBM, this powerful machine can execute up to 450 billion inference operations per day, representing a 50% performance leap from its predecessor, the IBM z16, launched in 2022.
Despite perceptions that mainframes are outdated, they remain mission-critical for many large enterprises—around 71% of Fortune 500 companies still rely on them. The global mainframe market was valued at approximately $5.3 billion in 2024, underscoring its continuing relevance.
IBM has positioned the z17 to seamlessly integrate with modern hardware ecosystems, open-source tools, and AI development environments. The system supports 48 IBM Spyre AI accelerator chips out of the box, with plans to expand to 96 chips within a year—ensuring scalability as new AI models and technologies emerge.
This innovation didn’t happen overnight. According to Tina Tarquinio, IBM’s VP of Product Management and Design for IBM Z, the z17 has been in development for over five years. The direction it took was heavily shaped by insights gathered from more than 100 enterprise customers during 2,000+ hours of research.
“What’s exciting,” said Tarquinio, “is how well that early customer feedback aligns with the AI-driven future we’re seeing unfold today. We knew we had to design for flexibility because the pace of AI innovation is only accelerating.”
One of the standout features of the z17 is its energy efficiency. IBM claims the system delivers 7.5x more AI acceleration while using 5.5x less energy than comparable multi-model platforms on the market.
The z17 is set for general availability starting June 8, offering enterprises a future-ready, AI-empowered solution that blends high performance, scalability, and sustainable design.w
