August 31, 2026 –
NVIDIA and MediaTek are significantly expanding their longstanding technology partnership, targeting the next generation of AI computing platforms across data-center infrastructure, local AI systems and automotive applications.
As part of the expanded collaboration, MediaTek will adopt NVIDIA NVLink Fusion as a foundation for developing custom AI accelerators, giving hyperscalers, cloud service providers and frontier AI model developers a prevalidated pathway to integrate custom XPUs with NVIDIA’s rack-scale AI infrastructure.
Alongside the technology partnership, NVIDIA has invested $3.5 billion in convertible bonds issued by MediaTek, strengthening the financial and strategic relationship between the two semiconductor companies.
The expanded collaboration combines NVIDIA’s expertise in accelerated computing, AI software and graphics with MediaTek’s capabilities in custom silicon, system-on-chip design, connectivity, advanced packaging and power-efficient computing.
Three Key Areas of Collaboration
The expanded NVIDIA-MediaTek partnership will focus on three major areas:
1. AI Infrastructure
MediaTek will work within the NVIDIA NVLink Fusion ecosystem to help customers develop custom AI infrastructure capable of integrating with NVIDIA rack-scale systems and AI factories.
The approach is designed to allow hyperscalers, cloud providers and AI companies to customise the compute components of their infrastructure while using NVIDIA technologies for connectivity, memory, packaging and system integration.
This could help organisations reduce the complexity involved in developing and deploying custom AI accelerators at scale.
2. Local AI Computing
NVIDIA and MediaTek will continue working across multiple generations of chips for local AI computing, including platforms for consumer PCs, AI developer systems and enterprise workstations.
The collaboration builds on MediaTek’s work on the GB10 Grace Blackwell Superchip, which powers NVIDIA DGX Spark. The platform combines an NVIDIA Blackwell GPU with a Grace CPU through NVLink-C2C connectivity.
The companies are also extending their work through NVIDIA RTX Spark platforms aimed at bringing powerful generative and agentic AI capabilities to consumer and professional computing devices.
3. Automotive AI
The partnership also extends into the automotive sector, where the two companies are developing platforms for AI-powered, software-defined vehicles.
MediaTek’s automotive system-on-chip expertise will be combined with NVIDIA’s accelerated computing, AI, graphics and software technologies.
MediaTek’s Dimensity Auto platforms integrate NVIDIA technologies to support advanced AI capabilities and NVIDIA RTX graphics for intelligent vehicle cockpits. These platforms can also operate alongside NVIDIA DRIVE AGX systems.
The companies plan to continue developing the architecture across future generations of automotive platforms.
What Is NVIDIA NVLink Fusion?
A major component of the expanded partnership is NVLink Fusion, NVIDIA’s platform for companies developing custom AI accelerators.
The platform provides a prebuilt and system-validated foundation for multi-die XPU development, helping customers move from chip design and advanced packaging toward rack-scale AI systems.
The technology stack includes several key components.
NVIDIA NVLink Fusion Chiplet connects custom XPUs to the NVIDIA NVLink scale-up fabric through photonics or electrical interconnects.
NVIDIA NVLink-C2C provides high-bandwidth, energy-efficient connections between XPUs, NVIDIA Grace CPUs and other compatible processors.
NVIDIA NVHBM supports customised memory architectures designed to increase bandwidth and energy efficiency while allowing more silicon area to be dedicated to computing.
Together, these technologies are intended to simplify the development of production-ready AI infrastructure.
Custom AI Accelerators Become More Important
Building a custom AI accelerator is only one part of creating a large-scale AI computing system.
Companies also need to integrate advanced packaging, high-speed SerDes, high-bandwidth memory, input/output technologies, networking and rack-scale infrastructure.
NVIDIA and MediaTek’s approach aims to reduce the engineering burden around these components.
Customers can focus on the computing architecture that differentiates their AI systems while relying on the broader NVIDIA and MediaTek ecosystem for connectivity, memory, packaging, manufacturing and rack-scale deployment.
MediaTek will also allow customers to tailor areas such as connectivity, memory, packaging, performance and power characteristics according to specific workloads.
NVIDIA and MediaTek Target the Next AI Computing Wave
The partnership reflects a broader shift in the semiconductor industry.
AI workloads are no longer limited to large cloud data centres. Generative AI and agentic AI are increasingly moving toward PCs, workstations, edge systems and vehicles, creating demand for different types of processors and architectures.
NVIDIA’s accelerated computing platform and software ecosystem gives the partnership access to a broad AI technology stack, while MediaTek brings expertise in efficient SoC design and connectivity across multiple computing categories.
For customers, the combination could create more options for deploying AI across cloud and edge environments.
Bringing AI Supercomputing Closer to the Edge
The companies’ local-computing collaboration is aimed at making high-performance AI available outside traditional data centres.
MediaTek’s contribution to the GB10 Grace Blackwell Superchip demonstrates how the companies are combining their technologies to create compact AI computing systems.
The next generation of RTX Spark platforms is expected to further extend this strategy into consumer PCs and professional systems.
This could become increasingly important as businesses and developers seek to run AI workloads locally for reasons including performance, privacy, latency and reduced dependence on cloud infrastructure.
AI Moves Into the Vehicle
Automotive computing represents another major growth opportunity.
Modern vehicles are increasingly becoming software-defined platforms, with AI being used for intelligent cockpits, driver assistance, infotainment and other vehicle functions.
By combining MediaTek’s automotive SoCs with NVIDIA’s computing and graphics technologies, the companies are positioning their partnership around the emerging physical AI market.
The collaboration is expected to continue across multiple generations of automotive platforms as vehicles become increasingly dependent on sophisticated computing architectures.
A Strategic Investment Beyond Semiconductors
NVIDIA’s $3.5 billion investment in MediaTek convertible bonds adds another dimension to the expanded relationship.
Beyond individual products, the partnership connects two major semiconductor capabilities across cloud AI infrastructure, edge computing and automotive technology.
For NVIDIA, MediaTek provides access to expertise in power-efficient SoCs, connectivity and custom silicon.
For MediaTek, the relationship provides an opportunity to integrate NVIDIA’s AI computing and software technologies into a broader range of platforms.
As AI increasingly becomes a fundamental layer across computing, the partnership could give both companies a stronger position across multiple parts of the technology ecosystem.
The Bigger Picture
The NVIDIA-MediaTek collaboration highlights how the next phase of AI computing is likely to extend well beyond conventional GPUs and hyperscale data centres.
Custom AI accelerators, local AI PCs and AI-defined vehicles are emerging as important parts of the market.
By combining NVIDIA’s accelerated computing and AI ecosystem with MediaTek’s custom silicon and SoC expertise, the companies are building a technology relationship that spans the cloud, edge and automotive sectors.
The $3.5 billion investment further underscores the strategic importance of that relationship as the global AI infrastructure market enters its next phase.
Key Takeaways
- NVIDIA has invested $3.5 billion in MediaTek convertible bonds.
- MediaTek will adopt NVIDIA NVLink Fusion for custom AI accelerator development.
- The partnership covers AI infrastructure, local AI computing and automotive technology.
- The companies are continuing work on AI-powered PCs and developer systems.
- NVIDIA and MediaTek are developing technologies for software-defined vehicles.
- NVLink Fusion is designed to simplify the integration of custom XPUs into NVIDIA rack-scale AI infrastructure.
- The partnership brings together expertise in accelerated computing, custom silicon, SoCs, connectivity and advanced packaging.
Ruchi Kumar is the associate editor at Entrepreneur News Network and TVW News India, where she leads editorial strategy, brand storytelling, and startup ecosystem coverage. With a strong focus on innovation, business, and marketing insights, he curates impactful narratives that spotlight India’s evolving entrepreneurial landscape. She has written extensively on fintech, AI and emerging startups.