scale up vize
scale up vize

Meta introduces new artificial intelligence training chip

ENGLISH 11.04.2024 - 10:46, Güncelleme: 11.04.2024 - 10:46
 

Meta introduces new artificial intelligence training chip

New version more than doubles compute, memory bandwidth
Meta introduced its new artificial intelligence (AI) training chip Wednesday that supports new generative AI products and services. The US-based global tech firm said its next generation Meta Training and Inference Accelerator, called MTIA, is part of the company's growing investment in its AI infrastructure that will complement its existing and future AI infrastructure to deliver new and better experiences across products and services. The company last year unveiled the MTIA v1, its first-generation AI inference accelerator that was designed in-house with Meta’s AI workloads. "This new version of MTIA more than doubles the compute and memory bandwidth of our previous solution while maintaining our close tie-in to our workloads," it said in a statement. "It is designed to efficiently serve the ranking and recommendation models that provide high-quality recommendations to users," it added.
New version more than doubles compute, memory bandwidth

Meta introduced its new artificial intelligence (AI) training chip Wednesday that supports new generative AI products and services.

The US-based global tech firm said its next generation Meta Training and Inference Accelerator, called MTIA, is part of the company's growing investment in its AI infrastructure that will complement its existing and future AI infrastructure to deliver new and better experiences across products and services.

The company last year unveiled the MTIA v1, its first-generation AI inference accelerator that was designed in-house with Meta’s AI workloads.

"This new version of MTIA more than doubles the compute and memory bandwidth of our previous solution while maintaining our close tie-in to our workloads," it said in a statement.

"It is designed to efficiently serve the ranking and recommendation models that provide high-quality recommendations to users," it added.

Habere ifade bırak !
Habere ait etiket tanımlanmamış.
Okuyucu Yorumları (0)

Yorumunuz başarıyla alındı, inceleme ardından en kısa sürede yayına alınacaktır.

Yorum yazarak Topluluk Kuralları’nı kabul etmiş bulunuyor ve turkishpress.co.uk sitesine yaptığınız yorumunuzla ilgili doğrudan veya dolaylı tüm sorumluluğu tek başınıza üstleniyorsunuz. Yazılan tüm yorumlardan site yönetimi hiçbir şekilde sorumlu tutulamaz.
Sitemizden en iyi şekilde faydalanabilmeniz için çerezler kullanılmaktadır, sitemizi kullanarak çerezleri kabul etmiş saylırsınız.