Nvidia and Intel unveiled a far-reaching partnership to co-develop custom chips for data centers and PCs, marrying Nvidia’s AI and accelerated-computing stack with Intel’s x86 platforms and manufacturing. Intel will design and produce Nvidia-custom x86 CPUs for use in Nvidia’s AI infrastructure, while also offering x86 PC system-on-chips that incorporate Nvidia RTX GPU chiplets via NVLink. Nvidia said it will purchase $5 billion of Intel common stock at $23.28 a share, a deal subject to regulatory approvals including under the Hart-Scott-Rodino Act. The collaboration underscores the industry’s rush to power AI workloads across cloud and client devices and could reshape competitive dynamics spanning CPUs, GPUs and chiplet-based designs. Both companies emphasized ecosystem breadth—CUDA and NVLink on Nvidia’s side, and Intel’s process tech, packaging and x86 software base—while cautioning that timelines and outcomes depend on development, market acceptance and regulatory clearance.
Related article:
— NVIDIA Unveils Rubin CPX: A New Class of GPU Designed for Massive-Context Inference
