AMD launches new accelerators and processors geared toward running large language models, or LLMs, for AI applications.

Chipmaker AMD (NASDAQ: AMD) has unveiled a range of innovative AI solutions spanning from data centers to personal computers. The AMD Instinct MI300 Series features data center AI accelerators, while the ROCm 6 open software stack introduces new features supporting Large Language Models (LLMs). Additionally, the Ryzen 8040 Series processors come equipped with Ryzen AI capabilities.

AMD’s CEO Lisa Su said the company was in a unique position to power the complete AI infrastructure, from large-scale cloud installations to enterprise clusters and AI-enabled devices. The demand for the Instinct MI300 GPUs, claimed to be the world’s highest-performing accelerators for generative AI, is reportedly strong. (Zee Business)

Lisa Su also highlighted partnerships with major cloud companies, leading server providers, and innovative AI startups, aiming to accelerate the adoption of Instinct MI300 solutions across the AI ecosystem. The AI market is projected to experience significant growth, with an expected compound annual growth rate of 70%, reaching USD 400 billion by 2027.

AMD’s Instinct MI300 Series accelerators leverage advanced technologies and are set to be deployed in large-scale cloud and enterprise environments. Customers include Microsoft (NASDAQ: MSFT), incorporating AMD Instinct MI300X accelerators in its Azure ND MI300x v5 Virtual Machine series for optimized AI workloads.

AMD’s advancements also extend to mobile processors, with the Ryzen 8040 Series offering robust AI compute capabilities.

The company unveiled Ryzen AI 1.0, a software stack facilitating easy deployment of AI-enabled apps for Windows.

Looking ahead, the next-gen ‘Strix Point’ CPUs, scheduled for a 2024 launch, will incorporate the AMD XDNA 2 architecture, promising over three times the AI compute performance compared to the prior generation.

Additionally, AMD introduced ROCm 6, an optimized open-source software stack for Instinct GPUs, particularly tailored for generative AI and large language models.

Related Story:  Microsoft Launches AI Chips To Power Its Azure Cloud