Meta releases information on a GPU cluster equipped with 24,576 NVIDIA H100 GPUs and used for training such as 'Llama 3'

'Leading AI development means leading in investment in hardware infrastructure,' says Meta, which is investing in data center scale with more than 24,000 GPUs. Revealed cluster information.

Building Meta's GenAI Infrastructure - Engineering at Meta

Meta reveals details of two new 24k GPU AI clusters - DCD

In 2022, it was reported that Meta will build an ``AI Research SuperCluster (RSC)'' equipped with 16,000 GPUs.

The new GPU cluster was built based on the lessons learned from this RSC, and is equipped with 24,576 NVIDIA H100 GPUs, which are said to be optimal for machine learning. Meta is believed to have purchased 150,000 NVIDIA H100 GPUs in 2023.

According to Meta, the new GPU cluster will be able to support larger and more complex models than RSC, paving the way for advances in generative AI development. In fact, this cluster will support current and next-generation AI models that Meta is working on, including the publicly available large-scale language model Llama 3, the successor to Llama 2, as well as generative AI and other It is said that it supports AI research and development in the field of.

This GPU cluster construction is just one step in Meta's infrastructure roadmap, which will include 350,000 NVIDIA H100 GPUs by the end of 2024, equivalent to 600,000 NVIDIA H100 GPUs in terms of computing power. The aim is to build a