Tech giants collaborate to establish standardized narrow precision data formats for AI.


October 18, 2023

The MX Alliance's efforts to standardize next-generation data formats for AI training and inference are driving the democratization of AI capabilities and paving the way for future innovations in AI modeling.

  • Optimizing every layer of the AI stack, from algorithms to hardware, is essential.
  • Narrower formats allow silicon to perform more efficient AI calculations per clock cycle, resulting in faster model training and inference times.
  • MX technology enables pre-training at 6- and 4-bit precisions without any modifications to conventional training recipes.


Efficiency and scalability are crucial for unlocking the full potential of next-generation deep learning. To achieve this, optimizing every layer of the AI stack, from algorithms to hardware, is essential. Recent advancements in narrow-precision AI data formats and optimized algorithms have played a significant role in this journey, allowing the industry to move from traditional 32-bit floating point precision to just 8 bits of precision.

By using narrower formats, silicon can perform more efficient AI calculations per clock cycle, resulting in faster model training and inference times. AI models also take up less space, reducing the need for data fetches from memory and improving performance and efficiency. Moreover, fewer bit transfers minimize data movement over the interconnect, leading to enhanced application performance and reduced network costs.

Recognizing the importance of standardization in this field, industry leaders including AMD, ARM, Intel, Meta, Microsoft, nVidia, and Qualcomm Technologies have come together to form the Microscaling Formats (MX) Alliance. Their goal is to create and standardize next-generation 6- and 4-bit data types for AI training and inferencing. The key enabling technology behind these sub 8-bit formats, known as microscaling, builds upon years of design space exploration and research. MX enhances the robustness and ease-of-use of existing 8-bit formats, making it easier for the industry to adopt single digit bit training and inference.

The initial MX specification introduces four concrete floating point and integer-based data formats (MXFP8, MXFP6, MXFP4, and MXINT8) that are compatible with current AI stacks. These formats provide implementation flexibility across both hardware and software and enable fine-grain microscaling at the hardware level. Extensive studies have shown that MX formats can be easily deployed in real-world applications such as large language models, computer vision, and recommender systems. Furthermore, MX technology enables pre-training at 6- and 4-bit precisions without any modifications to conventional training recipes.

In the rapidly evolving AI landscape, open standards are crucial to foster innovation, collaboration, and widespread adoption. These standards provide a unifying framework that ensures consistent toolchains, model development, and interoperability across the AI ecosystem. This empowers developers and organizations to fully harness the potential of AI while avoiding fragmentation and technology constraints that could hinder progress. In line with this philosophy, the MX Alliance has released the Microscaling Formats (MX) Specification v1.0 in an open, license-free format through the Open Compute Project Foundation (OCP). This move aims to encourage broad industry adoption and lay the foundation for potential future narrow-format innovations. In addition to the specification, a white paper and emulation libraries have been published to provide detailed insights into the data science approach and showcase select results of MX in action. This inclusive approach not only accelerates AI advancement but also promotes openness, accountability, and responsible AI development.

Founding members of the MX Alliance, such as AMD, Arm, Intel, Meta, Microsoft, NVIDIA, and Qualcomm Technologies, have expressed their support and commitment to this initiative. They believe that standardizing MX data formats will drive open and sustainable AI innovations while allowing the AI ecosystem time to prepare for future hardware and software implementations. These industry leaders emphasize the importance of collaboration and standardization to unlock the full potential of AI technology across various domains, from cloud to edge to endpoint. They believe that the OCP MX specification is a significant milestone in unifying and standardizing sub-8-bit data formats for AI applications, enabling higher levels of performance, energy efficiency, and reduced memory needs.

The Open Compute Project Foundation (OCP), responsible for fostering the OCP Community, is driving this initiative. The OCP Community comprises hyperscale data center operators, telecom providers, colocation providers, enterprise IT users, and the product and solution vendor ecosystem. Together, they develop open innovations that can be deployed from the cloud to the edge. The OCP Foundation aims to address market obstacles with open specifications, designs, and emerging market programs that showcase recognized IT equipment and data center facility best practices. Additionally, the organization invests in strategic initiatives and programs that prepare the IT ecosystem for major technology changes, such as AI and ML. The OCP Community’s open innovations strive to benefit all, focusing on impact, efficiency, scale, and sustainability.

The MX Alliance’s efforts to standardize next-generation data formats for AI training and inference are a significant step forward in the AI industry. By promoting interoperability and portability of AI models, these formats enable developers to focus on innovation while improving performance, energy efficiency, and memory usage. The release of the MX specification in an open format through the OCP Foundation fosters collaboration, accelerates AI advancements, and ensures responsible AI development. With the support of industry leaders, the MX Alliance is driving the democratization of AI capabilities and paving the way for future innovations in AI modeling.

Tech giants collaborate to establish standardized narrow precision data formats for AI.

(Source)

Background Information


About AMD: AMD, a large player in the semiconductor industry is known for its powerful processors and graphic solutions, AMD has consistently pushed the boundaries of performance, efficiency, and user experience. With a customer-centric approach, the company has cultivated a reputation for delivering high-performance solutions that cater to the needs of gamers, professionals, and general users. AMD's Ryzen series of processors have redefined the landscape of desktop and laptop computing, offering impressive multi-core performance and competitive pricing that has challenged the dominance of its competitors. Complementing its processor expertise, AMD's Radeon graphics cards have also earned accolades for their efficiency and exceptional graphical capabilities, making them a favored choice among gamers and content creators. The company's commitment to innovation and technology continues to shape the client computing landscape, providing users with powerful tools to fuel their digital endeavors.

AMD website  AMD LinkedIn

About ARM: ARM, originally known as Acorn RISC Machine, is a British semiconductor and software design company that specializes in creating energy-efficient microprocessors, system-on-chip (SoC) designs, and related technologies. Founded in 1990, ARM has become a important player in the global semiconductor industry and is widely recognized for its contributions to mobile computing, embedded systems, and Internet of Things (IoT) devices. ARM's microprocessor designs are based on the Reduced Instruction Set Computing (RISC) architecture, which prioritizes simplicity and efficiency in instruction execution. This approach has enabled ARM to produce highly efficient and power-saving processors that are used in a vast array of devices, ranging from smartphones and tablets to IoT devices, smart TVs, and more. The company does not manufacture its own chips but licenses its processor designs and intellectual property to a wide range of manufacturers, including Qualcomm, Apple, Samsung, and NVIDIA, who then integrate ARM's technology into their own SoCs. This licensing model has contributed to ARM's widespread adoption and influence across various industries.

ARM website  ARM LinkedIn

About Intel: Intel Corporation, a global technology leader, is for its semiconductor innovations that power computing and communication devices worldwide. As a pioneer in microprocessor technology, Intel has left an indelible mark on the evolution of computing with its processors that drive everything from PCs to data centers and beyond. With a history of advancements, Intel's relentless pursuit of innovation continues to shape the digital landscape, offering solutions that empower businesses and individuals to achieve new levels of productivity and connectivity.

Intel website  Intel LinkedIn

About Microsoft: Microsoft, founded by Bill Gates and Paul Allen in 1975 in Redmond, Washington, USA, is a technology giant known for its wide range of software products, including the Windows operating system, Office productivity suite, and cloud services like Azure. Microsoft also manufactures hardware, such as the Surface line of laptops and tablets, Xbox gaming consoles, and accessories.

Microsoft website  Microsoft LinkedIn

About nVidia: NVIDIA has firmly established itself as a leader in the realm of client computing, continuously pushing the boundaries of innovation in graphics and AI technologies. With a deep commitment to enhancing user experiences, NVIDIA's client computing business focuses on delivering solutions that power everything from gaming and creative workloads to enterprise applications. for its GeForce graphics cards, the company has redefined high-performance gaming, setting industry standards for realistic visuals, fluid frame rates, and immersive experiences. Complementing its gaming expertise, NVIDIA's Quadro and NVIDIA RTX graphics cards cater to professionals in design, content creation, and scientific fields, enabling real-time ray tracing and AI-driven workflows that elevate productivity and creativity to unprecedented heights. By seamlessly integrating graphics, AI, and software, NVIDIA continues to shape the landscape of client computing, fostering innovation and immersive interactions in a rapidly evolving digital world.

nVidia website  nVidia LinkedIn

About Qualcomm: Qualcomm, a leading American global semiconductor and telecommunications equipment company, has played a pivotal role in shaping the modern technology landscape. Founded in 1985, Qualcomm has been at the forefront of innovation, particularly in the realm of mobile communications and wireless technology. The company's advancements have been instrumental in the evolution of smartphones, powering devices with their Snapdragon processors that deliver exceptional performance and efficiency. With a strong commitment to research and development, Qualcomm has enabled the growth of 5G technology, paving the way for faster and more connected experiences.

Qualcomm website  Qualcomm LinkedIn



Leave a Reply