Meta to Deploy In-House Inference Chip Artemis For AI Upgrade

- Advertisement -
- Advertisement -

The new Meta chip, internally referred to as ‘Artemis’ is designed to perform the inference process,

Facebook parent, Meta, is set to upgrade its artificial intelligence (AI) capabilities with the deployment of a new, custom-designed chip within its data centres this year. This second-generation silicon, an advancement from the initial line Meta unveiled last year, is part of the company’s strategy to diminish its reliance on Nvidia’s widely used chips. This move could significantly impact Meta’s operational costs, particularly in running AI workloads crucial for the development of generative AI products across its platforms, including Facebook, Instagram, WhatsApp, and hardware devices like Ray-Ban smart glasses.

The new Meta chip, internally referred to as ‘Artemis’ is designed to perform the inference process, where models use their algorithms to make decisions and respond to user prompts. Meta is also reportedly working on a more ambitious chip capable of handling both training and inference tasks, similar to GPUs.

- Advertisement -

Despite the challenges encountered with the first generation of its Meta Training and Inference Accelerator (MTIA) program, the development of an inference chip by Meta could offer a more efficient solution for processing the company’s recommendation models than the currently used Nvidia processors. 

Artificial Intelligence Electronic Circuit. Microchip With Glowi
Image for representational purpose only

The implementation of Meta’s proprietary chip could lead to substantial savings, potentially reducing annual energy expenses by hundreds of millions of dollars and cutting down on chip purchasing costs by billions. This development reflects the broader industry trend of tech firms investing huge in the expense of chips, infrastructure, and energy for AI applications.

The company reportedly plans to launch the updated chip in 2024, stressing that it would complement the extensive array of off-the-shelf graphics processing units (GPUs) Meta is acquiring. This approach aims to achieve the optimal balance of performance and efficiency for Meta-specific workloads.

Meta’s CEO, Mark Zuckerberg, revealed plans to acquire approximately 350,000 Nvidia “H100” processors, in demand GPUs for AI, by the end of the year. This acquisition, along with contributions from other suppliers, is expected to provide Meta with the computing power equivalent to 600,000 H100s.

The decision marks a positive direction for Meta’s in-house AI silicon project, especially after the company had previously halted its first chip iteration in 2022 in favour of purchasing Nvidia’s GPUs. Nvidia’s GPUs are currently leading in the AI training process, which requires processing vast data sets to train models.

- Advertisement -

Related Artcles

LEAVE A REPLY

Please enter your comment!
Please enter your name here

Exclusive

Sameer Aggarwal, Founder and CEO, RevFin

“We Have Maintained A Net NPA Rate Below 1%!”

0
Maintaining a low non-performing asset (NPA) rate is considered one of the key factors determining the success of an NBFC (non-banking financial company) operating...

Six Steps To Building Better Business Assets

0
In the journey of asset-building, solving good problems trumps battling bad ones. Dive into six essential steps for building and leveraging effective assets—the game-changer...

India’s Solar Capacity Installation Down By 44.1%, Delayed Projects To Add To 2024 Installations

0
Large-scale solar installations saw a notable decline of 50.8% YoY to 5.8 GW from 11.7 GW in 2022, constituting 77.2% of the annual capacity...

Buzz

NTPC Starts Commercial Operation Of 70 MW Chhattargarh Solar Project

0
The project is designed to produce 370 million units of energy annually, which is equivalent to powering 60,000 households. NTPC Renewable Energy Limited (NTPC REL)...

Noida’s Vervesemi Announces Made-In-India ASIC

0
The newly developed chip, proudly made in India, targets domains such as weighing scales and Force Touch technologies. A fabless semiconductor company based in Noida,...
GlobalFoundries and Qualcomm

Qualcomm Doubles Investment In India, Plans To Invest $1Bn

0
Qualcomm has the largest number of engineers working out of India, with nearly 17,000 of its global workforce of around 40,000 based in the...

Important Sectors

Sameer Aggarwal, Founder and CEO, RevFin

“We Have Maintained A Net NPA Rate Below 1%!”

0
Maintaining a low non-performing asset (NPA) rate is considered one of the key factors determining the success of an NBFC (non-banking financial company) operating...

NTPC Starts Commercial Operation Of 70 MW Chhattargarh Solar Project

0
The project is designed to produce 370 million units of energy annually, which is equivalent to powering 60,000 households. NTPC Renewable Energy Limited (NTPC REL)...

Nasdaq Welcomes Lotus Tech With Modest Rise Post SPAC Deal

0
Lotus Technology's Nasdaq debut sees a 2% share increase post-merger with L Catterton-backed SPAC, marking a $7 billion valuation amidst a challenging EV market...

Log9, Trinity Cleantech Team Up for Interoperable Charging Network

0
They aim to lead the way in advancing Type 6 fast chargers, the next-gen standard for speed and convenience. Trinity's Thunder Platform will integrate...

BYD YangWang U9 Debuts at $236,000: 0-100 km/h in 2.36s, LFP Battery

0
BYD Co. launched its priciest vehicle, a fully-electric supercar costing 1.68 million yuan ($233,450), on Sunday, competing with gas-powered counterparts from brands like Ferrari...

Manufacturing

GlobalFoundries and Qualcomm

Qualcomm Doubles Investment In India, Plans To Invest $1Bn

0
Qualcomm has the largest number of engineers working out of India, with nearly 17,000 of its global workforce of around 40,000 based in the...

Applied Ventures Funds VVDN Technologies

0
The investment demonstrates Applied Ventures' dedication to promoting innovation, collaboration, and entrepreneurship within the semiconductor and electronics manufacturing sectors. Applied Ventures, LLC, the venture capital...

Haryana EV-CO To Manufacture L3, L5 EVs At New Plant

0
The EV maker is self-funded and operates a modern manufacturing facility with cutting-edge technology and equipment. It specializes in producing two and three-wheelers, with...
Sona Comstar

Sona Comstar: First Indian Auto Supplier With PLI Certification

0
Sona Comstar, a leading EV traction motor manufacturer in India, receives PLI certification for its electric two-wheeler traction motor. Sona BLW Precision Forgings, commonly known...

Indian Manufacturers To Get Chinese Mobile Production Orders

0
Dixon has secured an order from Oppo for making phones, with an order size of around 5-6 lakh units per month.  Chinese firms have finally...