Meta is building its first custom chip specifically for running AI models, the company announced Thursday. As Meta ramps up its AI efforts — CEO Mark Zuckerberg recently said the company sees “an opportunity to introduce AI agents to billions of people in a way that will be useful and meaningful” — the chip and other infrastructure plans unveiled Thursday, could be crucial tools for Meta to compete with other tech giants that are also investing significant resources in AI.
Meta’s new MTIA chip, which stands for Meta Training and Inference Accelerator, is the “in-house, custom accelerator chip family targeted at inference workloads,” Meta VP and head of infrastructure Santosh Janardhan wrote in a blog post. The chip apparently offers “more processing power and efficiency” than CPUs and is “adapted to our internal workloads”. With a combination of MTIA chips and GPUs, Janardhan said Meta believes “we will deliver better performance, less latency and greater efficiency for any workload.”
In addition to the MTIA, Meta is also introducing a new ASIC specifically to help transcode video, which it calls the “MSVP” or Meta Scalable Video Processor. It’s designed to support both “the high-quality transcoding required for VOD and the low latency and faster processing times that live streaming requires,” Meta said in a separate blog post, and “in the future” will help things like AI created content and AR and VR specific content for Meta’s apps.
Meta is also working on a “next-generation data center design” that will be “AI-optimized” and “faster and more cost-effective to build,” Janardhan said, and the company also praised the power of its Research SuperCluster (RSC) AI supercomputer, “which we believe is one of the fastest AI supercomputers in the world.” This isn’t exactly new rhetoric about Meta’s RSC; the company has been sharing a lot of praise for the supercomputer since last year. But as the company tries to differentiate itself from an increasing number of AI initiatives from many of the biggest technology giants – including other custom chips – it makes sense that Meta would want to brag about its confidence in its own AI hardware.