• MiPhi Semiconductors, in partnership with Intel and SUSE, launched a CPU-based solution that eliminates the need for GPUs in training small-scale AI models.
• The innovation optimizes CPU processing for AI workloads, making model training faster, more cost-effective, and accessible, especially for emerging economies.
• Collaboration empowers developers globally, combining MiPhi’s semiconductor expertise, Intel’s CPU architecture, and SUSE’s enterprise-grade Linux platform to democratize AI adoption.

MiPhi Semiconductors revealed the creation of a game-changing solution, in partnership with Intel and SUSE, to avoid the use of GPUs to train small-scale AI models. Developers and businesses can now utilize direct CPU upgrades to speed up the training of models, making AI development more convenient, efficient, and economical.

The solution solves one of the greatest challenges in the ecosystem of AI prohibitively expensive and scarce GPUs. By optimising the CPU processing of AI workloads through leading-edge hardware-software convergence, MiPhi, Intel, and SUSE are opening doors to new dimensions of scalability of AI training without massive infrastructure expenses.

On the announcement, Prasad Balakrishnan, MiPhi Semiconductors Pvt. Ltd, explained "Our partnership with Intel and SUSE demonstrates MiPhi's goal to democratize technology. By minimizing reliance on GPUs, we are making it possible for developers and organisations especially in emerging economies to embrace AI in a speedier, more economical manner. AI workloads are developing at an unparalleled velocity, and the underpinnings that drive them need to catch up". With this solution, CPUs are now able to tackle training tasks that were formerly the exclusive domain of GPUs, reducing expenses, streamlining deployment, and allowing businesses of every size to discover the full potential of AI free from the limitations of hardware.

This solution combines MiPhi’s semiconductor expertise, Intel’s CPU architecture and SUSE’s enterprise-grade Linux platform to deliver optimised AI training performance without reliance on dedicated GPUs. This cross-industry synergy empowers developers to train small-scale models including NLP, computer vision, and recommendation systems directly on CPUs with dramatically improved efficiency.

Also Read: Apple May Tap OpenAI or Anthropic to Supercharge Siri

"Intel is committed to bringing compute technologies that are powerful, efficient, and accessible. This partnership with MiPhi and SUSE further unlocks the promise of Intel CPU-based architecture for AI workloads. By optimizing performance at the silicon level and conforming to enterprise-ready software platforms, we are lowering barriers to AI adoption, and making it easier for developers and organizations to innovate at scale", said Santhosh Viswanathan, VP and MD, India Region, Intel.

SUSE has always been dedicated to developing open, dependable, and enterprise-ready platforms for the most stressful workloads. With this collaboration with MiPhi and Intel, we are empowering developers to speed up their innovation cycles, training AI models on safe and scalable Linux infrastructure without having to spend on costly GPUs. "In the process, we aim to democratise AI and enable industries to power their digital transformation journeys with increased speed", said Sachin Vig, Director Strategic Channels & Alliances Indian Subcontinent, SUS

The partnership reflects MiPhi, Intel, and SUSE's shared commitment to speeding up the adoption of AI all over the world through scalable, sustainable, and cost-effective solutions. This milestone not only represents a major milestone towards developing the next generation of AI infrastructure but also reflects India's growing position as an innovation powerhouse accelerating generational AI and semiconductor technologies.

Read More News :

Meta Taps US Contractors to Develop Hindi AI Chatbots

WhatsApp Tests Apple's Liquid Glass Design Ahead of iOS 26 Release