In today’s fast-paced business world, artificial intelligence (AI) is at the forefront of digital transformation. With businesses increasingly relying on AI-driven applications to gain a competitive edge, Red Hat is advancing the AI landscape with the release of Red Hat OpenShift AI 2.15, a powerful platform designed to scale AI and machine learning (ML) workloads across hybrid and multicloud environments.
Debuted at KubeCon + CloudNativeCon North America 2025, the latest update to Red Hat OpenShift AI introduces key innovations that will enhance how enterprises manage, deploy, and scale AI models. With AI investments reaching new heights, IDC projects that AI and automation will contribute $1 trillion in productivity gains by 2026. Red Hat’s robust platform is positioned to help businesses maximize the potential of their AI solutions while maintaining flexibility, security, and ethical integrity.
Unveiling New Features: OpenShift AI 2.15 Sets a New Standard
The Red Hat OpenShift AI 2.15 release introduces several advanced features, specifically designed to address the growing demands of enterprises adopting AI at scale. With an emphasis on performance, operational efficiency, and ethical AI practices, OpenShift AI 2.15 provides powerful tools to enhance the AI model lifecycle from development to deployment.
- Model Registry: One of the most notable additions in OpenShift AI 2.15 is the Model Registry, now available as a tech preview. This feature provides a centralized location for managing, versioning, and tracking AI models, whether they are generative or predictive. The registry not only improves collaboration across teams but also facilitates easier deployment and sharing of models across hybrid cloud environments. Moreover, Red Hat has contributed the Model Registry to the Kubeflow community, expanding the reach of this feature within the open-source ecosystem.
- Data Drift Detection: To ensure AI models remain accurate and reliable over time, OpenShift AI 2.15 introduces data drift detection. This new tool enables businesses to monitor discrepancies between the data used for model training and the data fed to models in real-world environments. By identifying shifts in data distributions, organizations can take proactive steps to ensure that models continue to produce trustworthy predictions and outcomes.
- Bias Detection Tools: Red Hat also integrates bias detection into OpenShift AI 2.15, leveraging tools from the TrustyAI open-source community. These tools allow enterprises to assess and monitor the fairness of their models, both during training and deployment. By ensuring that AI models are free from harmful biases, businesses can foster trust and transparency in their AI systems, an increasingly crucial consideration in today’s ethical AI landscape.
- LoRA-Based Fine-Tuning for LLMs: Large language models (LLMs), such as Llama 3, require substantial resources for fine-tuning. OpenShift AI 2.15 addresses this challenge by introducing LoRA-based fine-tuning, a technique that significantly reduces resource consumption while optimizing model performance. This new capability enables enterprises to scale AI workloads more efficiently, improving both performance and cost-effectiveness in cloud-native environments.
- Hardware Acceleration Support: OpenShift AI 2.15 also expands hardware support, introducing integration with NVIDIA NIM and AMD GPUs. The integration with NVIDIA NIM accelerates generative AI deployments by providing microservices that enhance inference performance across hybrid cloud environments. Meanwhile, the inclusion of AMD GPUs enables businesses to tap into alternative hardware options for more efficient training and inference, particularly for computationally heavy tasks.
Elevating AI Model Serving and Experimentation
As AI applications grow more sophisticated, enterprises require flexible, high-performance solutions for managing and serving models. OpenShift AI 2.15 delivers enhanced model serving capabilities, now supporting vLLM serving runtime for KServe, a tool that optimizes deployment of large language models. The runtime’s integration ensures that organizations can scale their AI deployments quickly while maintaining the flexibility to tailor configurations to specific needs.
In addition to model serving, OpenShift AI 2.15 introduces advanced experimentation capabilities to help data scientists fine-tune their workflows. With hyperparameter tuning via Ray Tune, data scientists can optimize their models for better accuracy while speeding up training times across distributed environments.
Furthermore, the update enhances data science pipeline management by improving experiment tracking. This allows teams to easily manage, compare, and analyze pipeline runs in a more organized manner, streamlining the process of model development and optimization.
Red Hat’s Vision for the Future of AI
As the demand for AI-enabled applications continues to grow, Red Hat remains committed to providing businesses with the tools they need to deploy AI at scale. The release of OpenShift AI 2.15 underscores Red Hat’s mission to deliver a platform that is both scalable and secure, while also offering flexibility in a variety of hybrid and multicloud environments.
Joe Fernandes, Vice President and General Manager of Red Hat’s AI business unit, shared:
“As enterprises explore the new world of capabilities offered by AI-enabled applications and workloads, we expect interest and demand for underlying platforms to increase as concrete strategies take shape. Enterprises must see returns on these investments via a reliable, scalable, and flexible AI platform that runs wherever their data lives across the hybrid cloud. The latest version of Red Hat OpenShift AI offers significant improvements in scalability, performance, and operational efficiency.”
With businesses increasingly depending on AI to drive critical processes, Red Hat’s approach ensures that enterprises can achieve long-term success without sacrificing flexibility or security. Red Hat’s dedication to innovation and ethical AI positions OpenShift AI 2.15 as an essential platform for companies looking to integrate advanced AI capabilities into their operations.
Additionally, Justin Boitano, Vice President of Enterprise AI Software at NVIDIA, emphasized the power of this collaboration, stating:
“Enterprises are seeking streamlined solutions to rapidly deploy AI applications. The integration of NVIDIA NIM with Red Hat OpenShift AI 2.15 enhances full-stack performance and scalability across hybrid cloud environments, helping development and IT teams efficiently and securely manage and accelerate their generative AI deployments.”
Looking Ahead: OpenShift AI’s Role in Shaping the AI Landscape
As enterprises continue to innovate and scale AI applications, Red Hat OpenShift AI 2.15 offers a robust platform that supports every phase of the AI lifecycle. With improvements to model management, ethical AI, and hardware integration, Red Hat is equipping organizations to meet the challenges of today’s AI-driven economy while ensuring that AI remains transparent, efficient, and reliable.
Set for general availability in mid-2025, OpenShift AI 2.15 is poised to be a critical tool for enterprises aiming to enhance operational efficiency, reduce costs, and innovate faster in an increasingly digital world.
For more details on OpenShift AI 2.15 and how to upgrade, visit Red Hat’s website
