NVIDIA dan MediaTek Perluas Kemitraan untuk Membangun Platform AI Edge-to-Cloud

NVIDIA dan MediaTek pada 31 Agustus 2026 mengumumkan perluasan kolaborasi lama mereka untuk membangun generasi berikutnya dari platform komputasi AI yang mencakup infrastruktur AI, komputasi AI lokal, dan otomotif. Sebagai bagian dari pengumuman tersebut, NVIDIA menyatakan telah menginvestasikan $3,5 miliar dalam obligasi konversi yang diterbitkan oleh MediaTek.

Berdasarkan perjanjian yang diperluas, MediaTek akan mengadopsi platform NVIDIA NVLink Fusion, yang oleh kedua perusahaan digambarkan sebagai jalur pra-divalidasi bagi hyperscaler, penyedia layanan cloud, dan pengembang model frontier untuk mengembangkan XPU khusus dan mengintegrasikannya ke dalam pabrik AI skala rak yang terhubung dengan NVIDIA NVLink. Kemitraan ini menggabungkan komputasi terakselerasi, AI, grafis, dan platform perangkat lunak NVIDIA dengan silikon khusus, komputasi berkinerja tinggi, desain sistem‑on‑chip yang hemat daya, kemasan canggih, interkoneksi, dan konektivitas MediaTek.

“AI mengubah setiap platform komputasi — mulai dari pabrik AI terbesar di dunia hingga PC dan mobil,” kata Jensen Huang, pendiri dan CEO NVIDIA, dalam pengumuman perusahaan. “MediaTek adalah salah satu perusahaan semikonduktor terkemuka di dunia, dengan keahlian luar biasa dalam desain sistem‑on‑chip, konektivitas, kinerja terdepan, dan efisiensi daya.”

Three Areas of Collaboration

The companies outlined three major areas of work.

In AI infrastructure, MediaTek will work within NVIDIA’s NVLink Fusion ecosystem to help customers develop custom AI infrastructure designed to integrate with NVIDIA rack-scale systems and AI factories.

In local AI computing, the companies will continue collaborating on multiple generations of NVIDIA RTX Spark and DGX Spark PC chips that integrate NVIDIA GPUs with MediaTek SoCs. These power consumer PCs, AI developer supercomputers and enterprise-class workstations.

In automotive, MediaTek and NVIDIA will continue developing platforms for AI-powered, software-defined vehicles in what the companies call the era of physical AI.

“MediaTek and NVIDIA share a vision for making advanced AI computing pervasive across the technology landscape,” said Rick Tsai, vice chairman and CEO of MediaTek. “NVIDIA’s investment strengthens a collaboration that spans cloud AI infrastructure, local AI computing and automotive in the era of physical AI.”

Custom AI Infrastructure With NVLink Fusion

MediaTek will offer the NVLink Fusion platform as a design foundation for customers building custom AI accelerators, allowing those platforms to evolve alongside future NVIDIA architectures. NVIDIA describes NVLink Fusion as a prebuilt, prequalified and system-prevalidated foundation for multi-die XPU development, intended to accelerate the path from silicon and advanced packaging to rack-scale systems.

According to NVIDIA, the platform brings together several technologies surrounding a custom XPU. The NVLink Fusion chiplet connects XPUs to the NVIDIA NVLink scale-up fabric using NVIDIA photonics or electrical interconnects. NVLink-C2C provides high-bandwidth, energy-efficient connectivity between XPUs, NVIDIA Rosa CPUs and other compatible processors. NVHBM integrates customized memory capabilities intended to increase bandwidth and energy efficiency while dedicating more silicon area to compute.

NVIDIA said that building a custom accelerator is only the first step in deploying custom XPUs in rack-scale AI factories, and that integrating multi-die architectures, advanced packaging, high-speed SerDes, HBM, I/O and scale-up networking into a production-ready system requires extensive chip-to-rack engineering, qualification and supply-chain support. Rather than engineering every element from the ground up, customers can focus on the differentiated compute that defines their platforms while relying on NVIDIA and MediaTek for NVLink connectivity, memory architecture, packaging, manufacturing and rack-scale technologies, the company said. Customers can bring XPU designs to MediaTek and tailor connectivity, memory, packaging, performance and power characteristics to their workloads.

MediaTek is also part of the broader NVIDIA NVLink Fusion ecosystem, which NVIDIA said enables hyperscalers, cloud service providers and frontier model developers to connect custom XPUs to its AI infrastructure. NVIDIA said partners can use its scale-up and scale-out technology stack and the NVIDIA MGX rack-scale architecture to reduce development complexity and accelerate time to market for semi-custom AI factories.

Local Computing and Automotive

On local computing, MediaTek collaborated with NVIDIA on the GB10 Grace Blackwell Superchip that powers NVIDIA DGX Spark, which combines an NVIDIA Blackwell GPU and Grace CPU connected by NVLink-C2C. The companies extended that work with NVIDIA RTX Spark for consumer PCs. MediaTek’s own account of the RTX Spark effort describes the pairing of its CPU and power-efficiency design with NVIDIA’s full-stack AI platform, and notes the first wave of RTX Spark-powered laptops is expected in Fall 2026, according to a MediaTek release.

In automotive, MediaTek Dimensity Auto platforms integrate NVIDIA technologies to deliver AI and NVIDIA RTX graphics for intelligent vehicle cockpits and can work alongside NVIDIA DRIVE AGX. The companies said they will continue building on this foundation across future generations.