AMD has unveiled its latest artificial intelligence (AI) processors in a move to challenge Nvidia's dominance in the AI hardware market.

The announcement was made at the Computex technology trade show in Taipei, where AMD CEO Lisa Su introduced the MI325X accelerator, an advanced version of the MI300 series, expected to be available in the fourth quarter of 2024 (via Reuters).

At the event, Lisa Su emphasized that AI is AMD's top priority. "We're at the beginning of an incredibly exciting time for the industry as AI transforms virtually every business, improves our quality of life, and reshapes every part of the computing market," the CEO added.

TAIWAN-TECH-BUSINESS-SEMICONDUCTORS-AI-COMPUTEX
(Photo : I-HWA CHENG/AFP via Getty Images)
Lisa Su, chairwoman and CEO of Advanced Micro Devices (AMD), delivers the opening keynote speech at Computex 2024, Taiwan's premier tech expo, in Taipei on June 3, 2024.

Everything AMD Announced

AMD's announcement of the MI325X is just the beginning. The company detailed its roadmap for developing AI chips over the next two years, introducing the MI350 series expected in 2025.

According to AMD, the MI350 will offer a 35-fold improvement in inference performance over the existing MI300 series. This leap in performance is aimed at meeting the growing demand for generative AI programs, which require advanced chips capable of handling complex data center applications.

Looking further ahead, AMD announced the MI400 series, set to launch in 2026. These chips will be based on a new architecture called "Next," marking a major shift in AMD's chip design.

AMD Expanding AI Capabilities

In addition to the MI325X and future MI series, AMD also introduced the Ryzen AI 300 series for AI laptops, designed to compete with Intel's Lunar Lake and Qualcomm's Snapdragon X.

These chips will power laptops equipped with Microsoft's AI chatbot Copilot, showcasing AMD's efforts to integrate AI capabilities across various devices.

At Computex, AMD also revealed its next-generation Zen 5 CPU architecture and XDNA 2 GPU NPU architecture. These will power the new AMD Ryzen 9000 series processors for gaming PCs and the third-generation Ryzen AI processors for ultrabooks.

The new NPU, which AMD calls the 'most powerful in the world' will enable up to 50 trillion operations per second (TOPs), facilitating advanced on-device AI experiences such as recall and live captioning.

Read Also: Microsoft Makes Massive $3.2 Billion AI Push in Sweden

AMD Goes Head-to-Head Against Nvidia

Nvidia currently holds nearly 80% of the AI semiconductor market, making it the clear leader in the industry. However, AMD is determined to close this gap. The competition between the two tech giants is heating up, especially as both companies commit to an annual release cycle for their AI chips.

Nvidia recently announced its next-gen AI chip, 'Rubin,' which is a successor to the Blackwell model. Nvidia CEO Jensen Huang confirmed that his company will release new AI chips yearly, a strategy AMD is now mirroring.

This new tempo is crucial as the market demands newer products and capabilities each year.

While AMD's shares remained flat following the announcements, Nvidia saw a 3% increase. Despite this, AMD's stock has more than doubled in value since the start of 2023, though this rise pales compared to Nvidia's seven-fold surge in the same period.

Related Article: Nvidia Unveils 'Rubin' Next-Gen AI Chip Platform for 2026 Rollout

Tech Times Writer John Lopez

ⓒ 2024 TECHTIMES.com All rights reserved. Do not reproduce without permission.
Join the Discussion