Asset Expiration Date: April 10,2029 15:32:00Asset Expiration Date: April 10,2029 15:32:00

AI Solutions

Dell Technologies Joins the PyTorch Foundation as Premier Member

Dell is committed to working with the open source community to make AI more accessible, efficient and impactful. Read more...
By   |  
    • Dell joins the PyTorch Foundation to boost open source AI innovation.
    • This move reinforces Dell’s dedication to scalable, open source AI solutions.
    • Our collaboration will focus on integrating PyTorch into secure IT ecosystems.

AI thrives on collaboration. At Dell Technologies, we believe in the power of an open ecosystem to drive human progress. That is why we are proud to announce that Dell has joined the PyTorch Foundation as a Premier Member.

This commitment enhances our contribution to the open source community and reinforces our influence in shaping the future of AI. By working alongside other industry leaders within the foundation, we can help accelerate the development of AI, making it more powerful and accessible for organizations everywhere. Our ultimate goal is to help organizations innovate with confidence and achieve their strategic goals through real-world applications of AI.

Advancing AI through open collaboration

Dell has long recognized the importance of PyTorch in the AI landscape. The Dell AI Factory provides customers with an end-to-end portfolio to build, deploy, and manage complete AI solutions. At the core of these solutions are advanced open source software frameworks like DeepSpeed. These tools are essential for customers using AI to gain efficiencies and solve their most complex business challenges.

Dell provides validated designs and fine-tuning guides, such as our work with torchtune on Dell AI infrastructure, to simplify the process for our customers. This hands-on experience demonstrates our deep understanding of the framework and its practical applications. Here are some key benefits of integrating these frameworks into our AI solutions.

    • Silicon diversity: PyTorch and vLLM create an abstraction layer, allowing AI models to run across a wide range of architectures including NVIDIA, AMD, Intel, Broadcom, and others. This aligns with Dell’s commitment to offer flexibility and choice for AI infrastructure, enabling innovation with the best tools available, today and in the future as we add more accelerators.
    • Efficient training and fine-tuning: DeepSpeed is used to accelerate distributed training and fine-tuning of large models, allowing Dell infrastructure to optimize GPU memory usage, increase throughput, and achieve faster iteration cycles. This enables multimodal models to benefit from efficient data and model parallelism, facilitating rapid communication and scaling across multiple GPUs and nodes.
    • Seamless portability: PyTorch is the leading AI model development framework, while vLLM has become the standard for high-performance AI inference. These frameworks are pivotal to providing an open source, on-premises stack for Dell AI Factory, enabling cloud-to-on-prem portability for AI proof-of-concepts.
    • Performance optimization: Significant performance optimizations are continuously implemented within PyTorch and vLLM for training and inferencing. These optimizations, such as asynchronous checkpointing, prefix caching, and disaggregated inferencing, enable us to proactively design our Dell AI Factory system architecture for optimal performance and efficiency.

Our premier membership in the PyTorch Foundation allows us to contribute more directly to these critical frameworks, fostering greater optimization and tighter integration between the hardware and software that powers AI. The result is a more seamless, efficient, and scalable experience for AI developers and end-users.

“Dell’s commitment to open source innovation is important to our goal of driving human progress. PyTorch, vLLM, and DeepSpeed are significant open source AI software frameworks within our Dell AI Factory solutions, which allow customers to gain efficiency in solving their business needs. Joining the PyTorch Foundation as a Premier Member further enables Dell to provide deeper optimization and integration that benefits AI developers, end users, and the open source community overall.”
— Ihab Tarazi, Chief Technology Officer and Senior Vice President at Dell Technologies, Infrastructure Solutions Group Core.

Building the future of AI, together

By becoming a Premier Member, we are not just supporting the PyTorch Foundation; we are actively investing in the future of the entire AI community. Dell powers the world’s most advanced AI systems with our cutting-edge server, storage, networking, and cyber resilience portfolio. Our global scale, extensive partner ecosystem, and innovation across hardware and software position us to make significant contributions to the open source community. This collaboration will lead to:

    • Enhanced Performance: Tighter integration between Dell’s AI Factory infrastructure and PyTorch will unlock new levels of performance and efficiency, allowing organizations to train and deploy models faster.
    • Greater Accessibility: By contributing to open standards, we help make advanced AI tools more accessible to a broader range of developers and researchers.
    • Increased Innovation: A stronger open source community fosters more rapid innovation, providing enterprises with the cutting-edge tools they need to maintain a competitive edge.

A commitment to enterprise-ready AI

For AI developers, IT decision-makers and business leaders, this membership signifies Dell’s focus on providing secure, scalable, and cost-effective AI solutions. We understand that adopting AI requires more than just powerful hardware; it demands a reliable, open, and supported ecosystem that integrates smoothly into existing IT environments.

By leveraging open frameworks, organizations gain access to cutting-edge tools and a robust community of developers, enabling faster innovation and reduced vendor lock-in. Dell continues to integrate open source technologies into enterprise-ready platforms, ensuring seamless deployment, scalability, security, and performance.

We are excited about this journey with the PyTorch Foundation and the open source community. Together, we can continue to push the boundaries of what is possible with AI and build a future where technology drives progress for everyone.

To learn more about how Dell is helping organizations accelerate their AI initiatives, explore the Dell AI Factory.

provided by submitter

About the Author: Onur Celebioglu

Onur Celebioglu is a Fellow, responsible for AI and HPC architecture and pathfinding in Dell’s ISG CTO group. Prior to that, he had been responsible for the design, development and integration of Dell’s HPC, AI and Data Analytics solutions and lead the Dell AI and HPC Innovation Lab. Onur has been part of the team that officially started the AI solutions engineering program at Dell and has over 25 years of industry experience in AI and HPC. His main areas of expertise are application performance analysis, high speed interconnects, parallel file systems and cluster provisioning tools. Onur has worked extensively in AI and HPC product development, published technical articles in various publications, conferences and journals. Onur has a BS in Electrical Engineering from METU, Turkey and a MS in Electrical and Computer Engineering from Carnegie Mellon University.