Infinitix and Phison Strategic Partnership for Enterprise AI Infrastructure

Infinitix, Taiwan’s developer AI infrastructure software provider and NVIDIA partner in the AI software domain, announced a strategic technology partnership with Phison Electronics Corp., global provider in NAND flash controllers and storage solutions.The collaboration integrates Phison’s aiDAPTIV+ intelligent storage technology with Infinitix’s AI-Stack enterprise management solution, creating a next-gen software-hardware co-designed architecture that effectively resolves memory bandwidth and scalability bottlenecks in enterprise generative AI training and inference.

As large language models continue driving explosive compute demand, enterprises struggle with limited GPU memory capacity. This joint solution extends GPU memory transparently via high-performance NAND storage, enabling larger models and longer-context workloads without exclusive dependence on costly HBM GPUs – delivering significantly better TCO and accelerating enterprise AI adoption.

“Today’s AI race is no longer won by compute power alone – it’s won by intelligent, unified resource orchestration,” said Wen-Yu Chen, CEO, Infinitix. “By bringing storage into the core AI infrastructure stack with Kubernetes-native integration, we give enterprises the flexibility, efficiency and cost advantages needed to build truly scalable AI factories.”

K.S. Pua, CEO, Phison, commented: “The paradigm has shifted from single-GPU heroics to holistic system-level innovation. aiDAPTIV+ transforms NAND from mere storage into a strategic tier of the AI memory hierarchy. Together with Infinitix, we’re setting a new standard for scalable, accessible, and sustainable AI infrastructure.”

Infinitix’s AI-Stack

Key Benefits of the Integrated Solution:

Kubernetes-native unified management of NVIDIA/AMD GPUs + aiDAPTIV+ storage nodes
Transparent GPU memory extension using AI-optimized NAND
Enablement of large-scale training and long-context inference at lower cost
Superior ROI and improved sustainability profile

This partnership reflects both companies’ commitment to re-architecting enterprise AI infrastructure for the hyperscale intelligence era. Further collaboration in intelligent storage, AI orchestration, and cloud-scale operations is already underway.