Meta is working on a new chip for AI

Meta is building its first custom chip specifically for running AI models, The company announced on 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 that will be useful and meaningful” — the chip and other infrastructure revealed Thursday The plans could be an important tool for Meta to compete with other tech giants investing significant resources in AI.

Meta’s new MTIA chip, which stands for Meta Training and Invention Accelerator, is its “in-house, custom accelerator chip family targeting enterprise workloads,” Santosh Janardhan, Meta VP and head of infrastructure, wrote in a blog post. The chip apparently offers “greater compute power and efficiency” than CPUs and is “optimized for our internal workloads”. With the combination of MTIA chips and GPUs, Janardhan said Meta believes “we will deliver improved performance, decreased latency, and greater efficiency for each workload.”

In addition to the MTIA, Meta is also introducing a new ASIC specifically to aid in video transcoding, which it calls the “MSVP,” or Meta Scalable Video Processor. It is designed to support “both the high-quality transcoding required for VOD as well as the low latency and fast processing times required for live streaming”. Meta said in a separate blog postAnd “in the future,” it will help bring AI-generated content and AR- and VR-specific content to 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,” Janardhan said, and the company also touted 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 from Meta about the RSC; The company has been praising this supercomputer a lot since last year. But as the company tries to stand up against many of the biggest technology giants — a growing number of AI initiatives, including other custom chips — it’s understandable that Meta would want to brag about its confidence in its AI hardware.