Market
Overview

ULTRARAM™

Applications

“The future of computing will be defined not only by faster processors, but by smarter memory architectures.”

Why Memory Matters

  • AI infrastructure investment accelerating globally.
  • Memory is becoming a key determinant of AI system performance.
  • Data movement is a major contributor to computing energy consumption.
  • Future AI systems will increasingly depend on new memory architectures.
  • Universal memory could help simplify tomorrow’s computing platforms.

Powering the Next Generation of AI Infrastructure, Secure Computing and Intelligent Systems

Artificial intelligence is reshaping the future of computing.

From large language models and AI data centres to autonomous robots, intelligent sensors and secure edge devices, AI workloads are growing in both scale and complexity. While processor performance continues to advance rapidly, the movement and storage of data are increasingly becoming the limiting factors in overall system performance, power consumption and cost.

Today’s memory hierarchy was developed for an era when computing workloads were fundamentally different. Modern AI systems continually transfer enormous volumes of data between processors and multiple layers of memory, creating latency, consuming significant amounts of energy and limiting the efficiency of increasingly powerful AI accelerators.

As AI infrastructure expands—from hyperscale data centres to autonomous edge systems—the need for new memory technologies has never been greater.

Quinas is developing ULTRARAM™, a new class of universal memory built upon patented quantum resonant tunnelling technology that combines DRAM-class speed with non-volatile storage and ultra-low energy operation in a single semiconductor device.

Our vision extends far beyond developing a faster memory device. We are building a semiconductor platform designed to enable the next generation of AI infrastructure, secure computing and intelligent systems.

Our long-term technology roadmap includes ULTRARAM-Neuro™, extending the ULTRARAM™ platform towards future compute-in-memory and neuromorphic AI architectures.

The following application areas illustrate where we believe ULTRARAM™ has the potential to deliver significant value as computing architectures continue to evolve.

ULTRARAM™

AI Infrastructure & Data Centres

Solving the Memory Bottleneck for Artificial Intelligence

Artificial intelligence is redefining the global computing industry.

The emergence of large language models, generative AI and increasingly autonomous AI systems has created unprecedented demand for computing performance. As AI workloads continue to grow, the limiting factor is no longer simply processor performance—it is increasingly the ability to move, access and store vast quantities of data efficiently.

Memory has become one of the defining challenges of the AI era.

Modern AI systems continuously transfer data between processors and multiple layers of memory. This movement of data consumes significant amounts of energy, introduces latency and increasingly limits the performance of advanced AI accelerators. As compute capability continues to improve, memory architecture is becoming a critical determinant of overall system efficiency.

This challenge extends across the entire AI ecosystem—from hyperscale cloud providers training foundation models to enterprise AI platforms and next-generation inference infrastructure.

Future AI infrastructure will increasingly depend upon memory technologies capable of delivering:

  • High-speed data access
  • Persistent data storage
  • Lower energy consumption
  • Higher system efficiency
  • Simplified memory architectures

These requirements are driving significant global investment in advanced memory technologies and heterogeneous computing architectures.

ULTRARAM™ has been developed to address these emerging challenges by combining DRAM-class speed with non-volatile storage in a single memory technology. Its unique combination of performance and persistence has the potential to simplify future memory hierarchies while reducing the energy associated with storing and moving data.

Looking further ahead, universal memory technologies may enable new system architectures in which memory becomes a more active component of the computing platform, supporting increasingly efficient AI infrastructure across both training and inference workloads.

Key Applications

  • Hyperscale AI data centres
  • AI training clusters
  • Large-scale inference platforms
  • AI accelerators
  • Heterogeneous computing architectures
  • High-performance server memory
  • Energy-efficient cloud infrastructure
  • Future AI memory subsystems

Why It Matters

The future of artificial intelligence will depend not only on faster processors, but also on fundamentally better memory.

As AI infrastructure continues to expand worldwide, innovations in memory technology will play an increasingly important role in determining the performance, efficiency and sustainability of next-generation computing platforms.

Quinas believes ULTRARAM™ can contribute to this transition by providing a new class of universal memory designed for the evolving demands of artificial intelligence.

“The future of AI will be determined not only by faster processors, but by smarter memory architectures.”

ULTRARAM™

Edge AI & Autonomous Systems

Intelligent Computing at the Point of Action

Artificial intelligence is rapidly expanding beyond the data centre.

While today’s largest AI models are trained within hyperscale computing infrastructure, the next generation of AI will increasingly operate where data is created—inside autonomous vehicles, industrial robots, intelligent cameras, medical devices, AI PCs and billions of connected edge devices.

This shift towards Edge AI is enabling systems to perceive, reason and respond in real time, reducing latency, improving privacy and minimising reliance on continuous cloud connectivity.

As AI becomes increasingly autonomous and agentic, memory will play an even more important role in enabling intelligent decision-making at the point of action.

Unlike hyperscale data centres, edge computing platforms operate within highly constrained environments. Energy consumption, thermal performance, physical size and system reliability are often more critical than raw processing power. Every milliwatt saved extends battery life, reduces cooling requirements and enables new classes of intelligent products.

Future Edge AI platforms will therefore require memory technologies capable of delivering:

  • High-speed access to frequently used data
  • Persistent storage without continuous power
  • Ultra-low energy operation
  • High endurance and long-term reliability
  • Compact, simplified system architectures

ULTRARAM™ has been developed with these future requirements in mind. By combining DRAM-class speed with non-volatile storage in a single memory technology, ULTRARAM™ has the potential to simplify future memory hierarchies while improving overall system efficiency.

The evolution of Edge AI is also driving increasing interest in compute-in-memory and neuromorphic architectures, where reducing the movement of data between memory and processors becomes fundamental to improving performance and lowering energy consumption. These long-term industry trends align closely with Quinas’ technology roadmap, including the ongoing development of ULTRARAM-Neuro™, which is exploring future memory architectures for next-generation AI systems.

Key Applications

  • AI PCs
  • Industrial automation
  • Robotics
  • Autonomous vehicles
  • Intelligent sensors
  • Smart manufacturing
  • Medical devices
  • Internet of Things (IoT)
  • Edge AI processors
  • Defence edge systems

As artificial intelligence continues to move from centralised cloud infrastructure towards billions of intelligent connected devices, memory will become an increasingly strategic component of overall system design. Quinas believes universal memory technologies such as ULTRARAM™ can help enable this transition while providing a foundation for future innovations in compute-in-memory, neuromorphic computing and intelligent edge systems.

ULTRARAM™

High-Performance Computing

Accelerating the World’s Most Demanding Workloads

High-Performance Computing (HPC) underpins many of the world’s most computationally intensive applications, from scientific research and engineering simulation to weather forecasting, pharmaceutical discovery and artificial intelligence.

As modern HPC systems increasingly integrate AI alongside traditional simulation and modelling, the demands placed upon memory continue to grow. Today’s processors and AI accelerators are capable of extraordinary computational performance, yet moving data efficiently between compute and memory remains one of the principal challenges limiting overall system performance and energy efficiency.

Memory is no longer simply a storage resource—it has become a strategic component of modern computing architecture.

Future HPC platforms will require memory technologies capable of delivering:

  • Extremely high-speed data access
  • Persistent storage without compromising performance
  • Improved energy efficiency
  • High endurance and reliability
  • Simplified system architectures

These requirements are driving significant innovation across advanced memory technologies, heterogeneous integration and next-generation semiconductor platforms.

ULTRARAM™ has been developed to address these emerging challenges by combining DRAM-class speed with non-volatile storage in a single memory technology. Its unique characteristics have the potential to simplify future memory hierarchies while reducing the energy required to store, retrieve and move data throughout increasingly complex computing systems.

As AI and HPC continue to converge, future computing platforms will increasingly benefit from memory technologies capable of supporting both high-performance numerical workloads and large-scale AI inference within more efficient system architectures.

Key Applications

  • Supercomputing
  • Scientific research
  • Engineering simulation
  • Climate and weather modelling
  • Pharmaceutical discovery
  • Digital twins
  • AI-enhanced HPC systems
  • Advanced research computing

As the boundaries between AI infrastructure and high-performance computing continue to disappear, innovations in memory technology will play an increasingly important role in enabling faster, more efficient and more sustainable computing platforms. Quinas believes universal memory technologies such as ULTRARAM™ can help support this evolution while contributing to the future of scientific discovery, industrial innovation and artificial intelligence.

ULTRARAM™

Cyber Security & Trusted Computing

Enabling the Next Generation of Secure AI Infrastructure

Artificial intelligence is becoming a foundational technology for governments, enterprises and critical infrastructure worldwide. From healthcare and financial services to energy, defence and public services, AI systems are increasingly responsible for processing sensitive information and supporting decisions where security, resilience and trust are paramount.

As AI becomes more deeply embedded within society, protecting data throughout its lifecycle is emerging as one of the defining challenges of modern computing.

Security is no longer solely a software challenge. The next generation of computing platforms will increasingly depend upon trusted hardware, confidential computing environments and resilient system architectures that work together to protect sensitive information from the edge to the cloud.

Memory plays a central role within these systems.

Every AI workload depends upon memory to store, retrieve and process information efficiently. As confidential computing and privacy-preserving AI continue to evolve, future memory technologies will become an increasingly important part of secure computing platforms.

ULTRARAM™ is being developed as a new class of universal memory combining DRAM-class speed with non-volatile storage and ultra-low energy operation. While its primary objective is to enable more efficient computing, these characteristics may also support future trusted computing architectures where performance, persistence and energy efficiency are all critical design considerations.

The rapid growth of confidential computing, trusted AI, sovereign digital infrastructure and cyber resilience is creating demand for innovative semiconductor technologies capable of supporting increasingly secure computing platforms. Quinas believes universal memory technologies such as ULTRARAM™ have the potential to contribute to this long-term evolution.

Key Applications

  • Confidential computing platforms
  • Trusted AI infrastructure
  • Secure edge computing
  • Privacy-preserving AI
  • Critical national infrastructure
  • Government digital services
  • Secure cloud computing
  • Cyber-resilient AI systems

As artificial intelligence becomes increasingly integrated into critical digital infrastructure, trusted hardware will play an ever more important role in delivering secure, resilient and energy-efficient computing. Quinas is developing ULTRARAM™ as part of a next-generation semiconductor platform designed to support the future of AI infrastructure, secure computing and intelligent systems.

ULTRARAM™

Mission-Critical Computing

 
Enabling Intelligent Systems Where Performance, Reliability and Resilience Matter Most
 
The world’s most demanding computing environments operate under conditions where reliability is essential and failure is not an option.
 
From satellites orbiting the Earth and autonomous aerospace systems to defence platforms, secure communications and critical national infrastructure, mission-critical computing requires technologies capable of delivering exceptional performance while operating within strict power, environmental and reliability constraints.
 
Artificial intelligence is accelerating this transformation.
 
Modern mission-critical systems increasingly rely upon AI to improve situational awareness, automate decision making, process sensor data and support autonomous operations. As these systems become more intelligent and interconnected, memory becomes an increasingly important component of overall system architecture.
 
Unlike conventional commercial computing, mission-critical platforms often operate in environments where energy is limited, maintenance opportunities are infrequent and operational resilience is paramount.
 
Future computing platforms will increasingly require memory technologies capable of delivering:
 
  • High-speed access to critical data
  • Persistent storage without continuous power
  • Ultra-low energy operation
  • High endurance and long operational lifetimes
  • Reliable operation across demanding environments
  • Simplified system architectures
 
ULTRARAM™ is being developed as a new class of universal memory combining DRAM-class speed with non-volatile storage and ultra-low energy operation. These characteristics have the potential to support future mission-critical computing platforms where performance, persistence and efficiency are all essential.
 
As governments and industry invest in sovereign semiconductor capability, resilient AI infrastructure and advanced digital systems, innovative memory technologies will become increasingly important in enabling the next generation of secure, intelligent computing platforms.
 

Key Applications

 
  • Space Systems
 
Future satellites, Earth observation platforms, space robotics and onboard spacecraft computing demand highly efficient memory technologies capable of operating reliably throughout long-duration missions while minimising power consumption and system complexity.
 
  • Aerospace
 
Advanced avionics, autonomous aircraft and next-generation aerospace systems increasingly depend upon intelligent computing platforms capable of processing large volumes of data safely and efficiently.
 
  • Defence Systems
 
Artificial intelligence is becoming central to defence technologies, supporting autonomous systems, sensor fusion, secure communications and mission planning. Future defence platforms will require increasingly efficient memory architectures capable of supporting these demanding workloads.
 
  • Critical National Infrastructure
 
Energy networks, transportation systems, telecommunications infrastructure and public services are becoming increasingly dependent upon secure, resilient AI-enabled computing.
 
  • Sovereign Technology
 
Governments around the world are investing in trusted semiconductor capability, resilient supply chains and sovereign AI infrastructure to strengthen national technological resilience.
 
As intelligent systems become increasingly embedded within critical infrastructure, space platforms, aerospace technologies and sovereign digital capability, memory will continue to play a fundamental role in determining overall system performance, resilience and efficiency.
 
Quinas believes universal memory technologies such as ULTRARAM™ have the potential to support this long-term evolution by enabling more capable, energy-efficient and resilient computing platforms for some of the world’s most demanding applications.

ULTRARAM™

Future Computing Architectures

Building the Semiconductor Platform for the Next Generation of Intelligent Systems

Artificial intelligence is driving the most significant transformation in computing architecture since the invention of the microprocessor.

For more than half a century, advances in computing have been driven primarily by improvements in processor performance. Today, however, the rapid growth of AI is reshaping the entire computing stack. Increasingly, system performance, energy consumption and scalability are determined not only by processing capability, but by how efficiently data can be stored, accessed and moved throughout the system.

This shift is accelerating global investment in advanced memory technologies, heterogeneous semiconductor integration, chiplet architectures and novel computing platforms designed to overcome the limitations of today’s processor-memory hierarchy.

Quinas believes universal memory will become one of the defining technologies of this new era.

ULTRARAM™ is being developed as a new class of universal memory that combines DRAM-class speed with non-volatile storage and ultra-low energy operation in a single technology. While its immediate focus is enabling more efficient memory architectures, its long-term opportunity extends much further.

Future computing platforms are expected to integrate memory and processing more closely, reducing the movement of data, improving overall system efficiency and enabling increasingly intelligent computing architectures. This evolution is driving growing interest in technologies including compute-in-memory, neuromorphic computing, heterogeneous semiconductor integration and advanced AI hardware.

As part of its long-term technology roadmap, Quinas is developing ULTRARAM-Neuro™, extending the principles of ULTRARAM™ towards future compute-in-memory and neuromorphic architectures where memory becomes an increasingly active participant in computation rather than simply a location for storing data.

Emerging technologies—including advanced AI accelerators, quantum computing and heterogeneous semiconductor platforms—will continue to increase demand for innovative memory architectures capable of supporting the next generation of intelligent systems.

Future Technology Roadmap

  • Universal memory architectures
  • Compute-in-memory systems
  • ULTRARAM-Neuro™
  • Neuromorphic AI hardware
  • Heterogeneous semiconductor integration
  • Advanced chiplet architectures
  • Emerging quantum computing platforms

ULTRARAM™

One Platform. Multiple Generations of Computing

Quinas is not simply developing another memory device.

We are building a semiconductor platform designed to support the future of artificial intelligence, secure computing and intelligent systems.

Today, that platform is centred on ULTRARAM™, addressing the growing demand for high-performance, energy-efficient universal memory across AI infrastructure, edge computing, high-performance computing and mission-critical applications.

Tomorrow, our roadmap extends towards ULTRARAM-Neuro™, compute-in-memory and future semiconductor architectures that bring memory and processing closer together, enabling increasingly intelligent and energy-efficient computing systems.

As the boundaries between memory, processing and artificial intelligence continue to converge, universal memory has the potential to become a foundational technology within the next generation of semiconductor platforms.

Quinas is committed to advancing this vision through continued innovation, strategic partnerships and global collaboration with semiconductor manufacturers, foundries, research organisations, governments and technology leaders.

Together, we are helping to build the future of intelligent computing.

 
The future of computing will be shaped not only by faster processors, but by smarter semiconductor and memory architectures. Through ULTRARAM™ and ULTRARAM-Neuro™, Quinas is building a next-generation semiconductor platform to help power the future of AI infrastructure, secure computing and intelligent systems.