Meta, the company behind Facebook, Instagram, and WhatsApp, is gearing up to deploy its very own custom-designed artificial intelligence chips named Artemis. This move is expected to happen this year and could have significant implications for Meta‘s operations.
The American tech company has been spending huge sums of money to expand its computing capacity to support AI-driven features on its platforms. These features, like personalized recommendations and automated responses, require powerful hardware. By developing Artemis, Meta aims to reduce its reliance on chips from Nvidia, a leading supplier in the market.
According to Dylan Patel, a tech expert, using Artemis could save Meta millions in energy bills and billions in chip purchases. This is because Artemis is designed to be more energy-efficient and cost-effective for Meta’s specific needs.
Continued Partnership with Nvidia
Despite creating its own chip, Meta isn’t cutting ties with Nvidia completely. CEO Mark Zuckerberg has confirmed that the company will still use Nvidia’s chips alongside Artemis. In fact, Meta plans to have a significant number of Nvidia processors in its data centers by the end of the year.
Meta’s journey towards creating its own chip hasn’t been smooth. In 2022, the company had to abandon its first attempt and opt for Nvidia’s GPUs instead. However, the development of Artemis marks a positive step forward for Meta’s in-house chip project.
What Artemis Does?
Artemis, like its predecessor, is designed for AI inference tasks. This means it’s optimized for tasks like ranking content and generating responses based on user inputs. Meta believes that Artemis, combined with Nvidia’s GPUs, will provide the best performance and efficiency for its AI needs.
While Meta’s move to develop its own chip could shake up the AI market, it’s clear that Nvidia’s GPUs will still be a crucial part of Meta’s infrastructure for the foreseeable future.