"Rather than placing a bet and waiting for a long period of time, we deliberately take an iterative approach. Each MTIA generation builds on the last, using modular chiplets and incorporating the latest AI workload insights and hardware technologies."
"AI models are evolving faster than traditional chip development cycles, so AI workloads may change substantially by the time new hardware typically reaches production. Putting out fresh silicon this quickly is unusual by most chip industry standards, but is essentially unheard of for a social media company."
"The MTIA 300 will be used primarily for training algorithms that rank and recommend content to the hundreds of millions of people who use apps like Facebook and Instagram each day. The other three chips are designed to support inference, the process of running trained AI models."
Meta announced four new computer chips built on RISC-V architecture to power generative AI features and content ranking systems across its apps. Developed in partnership with Broadcom and manufactured by TSMC, these chips extend Meta's MTIA line. The MTIA 300 is already in production for training content ranking algorithms, while the MTIA 400, 450, and 500 are designed for inference workloads and expected to ship between early and late 2027. Meta's iterative approach to chip development addresses the rapid evolution of AI models, allowing each generation to incorporate latest AI insights and hardware technologies rather than waiting for traditional long development cycles.
#ai-chips #meta-hardware #mtia-accelerators #semiconductor-development #generative-ai-infrastructure
Read at WIRED
Unable to calculate read time
Collection
[
|
...
]