October 7, 2025 | By: Bargav Balakrishnan, VP, Product Management
Today, IBM® Power® is announcing a series of innovations that include one click to install1 and one configuration2 to move AI services from the IBM-supported catalog across Power processor-based servers on-prem and in the cloud on Power Virtual Server, enabling a turnkey solution for AI enterprise workflows. With these innovations, we are continuing to help enterprises accelerate their hybrid and modernization journeys while enhancing cyber resiliency to mitigate risks associated with ransomware attacks and quantum computing. These enhancements build on the introduction of Power11, our latest generation of servers designed to deliver autonomous IT for the AI era, announced on July 8th
Deploy AI across business workflows
A recent IBM IBV report showed that among companies launching AI pilots last year, over 50% of IT decision makers say that half or fewer of those pilots will have fully launched by the end of the year.[i] For organizations to be successful, they need to learn from feedback, retain context, and customize deployments deeply to specific workflows – starting at the workflow edges then scaling into core processes.
IBM Power is focused on enabling integration of core data and applications into AI workflows and business processes. We are announcing support for a full stack solution, from on- and off-chip acceleration with MMA and IBM Spyre™, to integrated and optimized AI inferencing with Red Hat AI Inference Server and OpenShift AI, and data services orchestration with watsonx.data®, that along with pre-built AI services and proven adoption patterns can help clients scale enterprise AI.
IBM Spyre Accelerator for Power: IBM Spyre enables an AI turnkey solution for AI enterprise workloads that is anchored on simplicity and speed. It brings off-chip acceleration that offers scalable capabilities for generative AI use cases on IBM Power processor-based servers where business-critical back-office workflows and applications are running. Lars Nylund, CEO of BanFast Forvaltning AB, said “With IBM Spyre Accelerator for Power, we can host our digital AI assistant locally in Sweden. This allows us to serve a wide range of clients, including the ones who demand the highest data protection standards and compliance with European and national regulations.” Other Retrieval-Augmented Generation (RAG) use cases for IBM Spyre Accelerator on Power will include digital assistants in IT operations, agriculture, real estate, and government with private documents. It will be available on December 12th, 2025*.
Red Hat AI Inference Server: Built on technologies like vLLM and model compression, Red Hat AI Inference Server streamlines deployment of large language models for high throughput, low latency, and cost-efficient performance across hybrid cloud environments. Running it on Power processor-based servers amplifies these benefits with high availability, advanced security features, and hardware-accelerated inferencing, built to helping organizations reduce latency, increase efficiency, and scale AI workloads with confidence. For clients using AI on Power, this means a consistent, enterprise-ready foundation for bringing generative AI into core business processes.
Red Hat OpenShift AI: Red Hat OpenShift AI on Power delivers a flexible, end to end enterprise-ready platform for the AI lifecycle, from data preparation and model development to deployment, serving, and monitoring, built on top of Red Hat OpenShift, a leading hybrid cloud application platform powered by Kubernetes. By running OpenShift AI on Power, clients gain the combined advantage of building a trustworthy AI operations environment with the performance, scalability, and resiliency of Power processor-based servers. This pairing helps organizations accelerate innovation while driving enterprise-grade reliability and security capabilities, making it easier to move from experimentation to production with confidence. For businesses adopting AI on Power, the results can include faster time to insight, address risk, and a consistent hybrid cloud platform that supports both predictive and generative AI workloads across industries.
IBM watsonx.data on Power: IBM is enabling enterprises to advance their AI journeys by combining the flexibility of an open, hybrid data lakehouse with the performance and resiliency of Power processor-based servers. Built to support generative AI and modern analytics, watsonx.data on Power allows organizations to unify, enrich, and govern both structured and unstructured data at scale, providing insights for AI agents and business applications. It will be available on December 10, 2025*.
Enhancing hybrid cloud, modernization, and automation
Power11 is designed to deliver the availability, resiliency, performance, and scalability enterprises demand, for an agile hybrid deployment on-premises or in the IBM Cloud. This helps enable clients to deploy applications and workloads where they are best fit to optimize for cost and performance while prioritizing enterprise-grade security and reliability. The following announcements are delivering innovation that can adapt to evolving business demands:
- IBM Power Virtual Server: Organizations can now take advantage of watsonx integration toolkits that make it easier than ever to connect Power Virtual Server-based applications and data to co-located watsonx SaaS and IBM Cloud services without the complexity of building or managing their own AI environments. We have also introduced Shared Processor Pools optimization to help improve efficiency and reduce licensing costs for AIX and IBM i applications and workloads, along with expanded IBM i Software Tiers that provide greater flexibility and cost alignment through Virtual Serial Number capabilities. Additionally, IBM Power Cyber Vault capability in collaboration with Celerity’s CopyAssure allows organization to gain enhanced protection against cyberattacks through automated clean room services.
- IBM i: We are announcing a broad set of October updates that enhance application development, modernization, and performance across the platform. This includes the latest Technology Refreshes IBM i 7.6 TR1 and IBM i 7.5 TR7, which deliver new system administration features, SQL enhancements, performance improvements, and high availability capabilities. We are also introducing IBM Rational Developer for i V9.9, bringing modern IDE tooling with expanded language support and Git integration, along with new Enterprise Support for IBM i Extensions in Visual Studio Code, giving developers more choice and flexibility with open source-friendly tools. Together these advancements reaffirm IBM i on Power as an IBM strategic platform for innovation, helping clients modernize applications, strengthen security, and confidently scale for the future.
- IBM Power Cyber Vault: IBM Power Cyber Vault is designed to deliver comprehensive cyber resiliency to enterprise clientsfor proactive protection,near real-time detection, automated response, and rapid recovery from ransomware threats. As part of our continued commitment to enhancing client experience and advancing IBM Power Cyber Vault capabilities, we are introducing key innovations in Q4, including less than one minute ransomware threat detection4 for IBM i with IBM FlashSystem® and FCM4 and enhanced threat detection accuracy through the newly released IBM PowerSC multi-layer correlation feature. IBM Power Cyber Vault is now available across AIX, IBM i, and Linux.
Since the launch of Power11 in July, our ISV ecosystem has rapidly embraced the platform, delivering transformative solutions across application modernization, AI, data, and hybrid cloud. Graph database solution provider Rocketgraph helps customers address revenue losses by detecting complex patterns like fraud in datasets which can be too large for standard SQL queries, and on Power11 clients can execute queries running Rocketgraph Accelerator 2.4 3.37X faster than on comparative x86 processor-based servers3. ISVs like Crest and Clari5 are leveraging Spyre to help clients protect against data breaches and fraudulent transactions. Met Rogers, SVP SRE at mission-critical banking ISV FIS shared, “As a long-standing partner of IBM, FIS is excited by the industry-leading innovation in the new Power11 systems. We look forward to harnessing next-generation capabilities, with innovations across the processor, system, and stack. We are confident that Power11 will empower financial institutions to fast-track their digital transformation journeys and fortify their mission-critical operations.” Partnering with leading ISVs, Power11 is unlocking scalable, real-world solutions that can turn innovation into tangible business outcomes for our clients.
By combining next-generation AI capabilities with enterprise-grade performance and resiliency, Power enables organizations to accelerate their modernization journeys, unlock new business value, and build an intelligent foundation for the challenges of today and tomorrow. Together, these innovations affirm Power as a foundation for enterprise transformation in an AI-driven world.
* IBM's statements regarding its plans, directions, and intent are subject to change or withdrawal without notice at IBM's sole discretion. Information regarding potential future products is intended to outline our general product direction and it should not be relied on in making a purchasing decision. The information mentioned regarding potential future products is not a commitment, promise, or legal obligation to deliver any material, code, or functionality. Information about potential future products may not be incorporated into any contract. The development, release, and timing of any future features or functionality described for our products remains at our sole discretion.
i https://filecache.mediaroom.com/mr5mr_ibmnewsroom/198550/IBM_ROI_of_AI_Report-December_2024.pdf
1 AI service of the IBM-supported catalog is delivered as one or a set of containers that can be deployed with a single deployment command. The provided UI for the catalog executes such commands in the backend based on a single click within the UI page of the respective AI service.
2 Single configuration enabled by exposed industry standard APIs to decouple services at the top and the backing inferencing service for all AI services that are part of the IBM-supported catalog. Any service that requires AI inferencing capabilities can connect inferencing services that provide OpenAI API or watsonx.ai API compliant inferencing endpoints (Spyre endpoint, RH AI Inferencing Server, IBM Cloud, OpenAI, Azure, AWS, GCP, ...). Services can run either on IBM Power or on IBM Power Virtual Server.
3 Based on Rocketgraph internal testing of Rocketgraph Accelerator 2.4 query response time for graphs from 2 to 3.9 billion netflow edges. Comparison is based on IBM Power E1180 (8 cores/2TB) running Linux 9.4 versus Intel Xeon x86 (8 cores/2TB) systems running Rocky Linux 9. Results valid as of Sept. 25, 2025, and conducted under laboratory conditions, individual results can vary based on workload size, use of storage subsystems and other conditions.
4 This guarantee covers only the displaying of an alert in less then one minute. Remediation is in the form of drive replacement up to the cost of the Covered Product. Terms and conditions apply, full details can be found here