IBM z17: The first mainframe fully engineered for the AI age

IBM has announced the z17 mainframe, engineered to support Artificial Intelligence (AI) workloads across enterprises. Featuring the IBM Telum® II processor, the z17 significantly enhances AI capabilities, allowing for real-time transaction scoring and an increase in AI inferencing operations.

Designed based on input from over 100 clients, the z17 boasts multi-model AI capabilities and improved security features. With the ability to perform more than 450 billion AI inference operations daily, it addresses various use cases, such as loan risk and chatbot management.

Ross Mauri, general manager of IBM Z and LinuxONE, stated, “The industry is quickly learning that AI will only be as valuable as the infrastructure it runs on. With z17, we're bringing AI to the core of the enterprise...” Further advancements include the IBM Spyre™ Accelerator for enhanced AI compute capabilities and new tools aimed at improving user experience through AI assistants.

The system integrates tightly with hybrid environments and is set to support advanced operational AI insights through the new z/OS 3.2 Operating System (OS), planned for release in Q3 2025. IBM Z Operations Unite further streamlines operations, combining metrics from various sources for more efficient incident response.

Additionally, IBM emphasizes security enhancements with AI-driven data protection methods and secrets management capabilities through HashiCorp's IBM Vault. Clients can rely on tailored support for mission-critical operations, reflecting IBM's continued focus on optimizing performance and security.

General Availability (GA) of the IBM z17 is slated for June 18, 2025, with the Spyre™ Accelerator expected in Q4 2025.