NVIDIA launches Ising AI model as Meta scales AI infrastructure

NVIDIA (NASDAQ: NVDA) recently launched what it described as the world’s first open artificial intelligence (AI) model to deliver breakthrough performance for quantum computers, while Meta (NASDAQ: META) is collaborating with chip designer Broadcom to power its AI ambition.

NVIDIA rolls out Ising—a breakthrough for AI-based quantum processor

In a press release published by NVIDIA on April 14, the tech giant announced that NVIDIA Ising was made to “help researchers and enterprises build quantum processors capable of running useful applications.” Named after a significant mathematical model that simplifies the analysis of complex physical systems, the NVIDIA Ising aims to provide high-performance AI tools for quantum error correction and calibration.

The Ising model would enable researchers to tackle much larger and more complex problems with quantum computers by delivering up to 2.5x faster performance and 3x better accuracy during the decoding process, NVIDIA’s press release said. It includes state-of-the-art customizable models, tools, as well as data such as Ising Calibration and Ising Decoding, to improve the use of quantum processors.

With the Ising Calibration, users and AI agents can quickly interpret and react to measurements from quantum processors, enabling automating continuous calibration and, as a result, reducing the time needed from days to hours. While the Ising Decoding optimized speed and accuracy by performing real-time decoding for quantum error correction. It is up to 3x more accurate than pyMatching, the current open source industry standard.

“AI is essential to making quantum computing practical,” Jensen Huang, founder and CEO of NVIDIA, expressed. “With Ising, AI becomes the control plane — the operating system of quantum machines — transforming fragile qubits to scalable and reliable quantum-GPU systems.”

Meta powers its AI ambition with a new deal

In other news, Meta has announced it will work with chip designer Broadcom. The collaboration aims to produce custom AI processors under an expanded deal as the company builds its computing capacity.

The April 14 announcement extends the deal until 2029 and includes an initial commitment of one gigawatt of computing capacity. Also, as part of the deal, Broadcom CEO Hock Tan would leave Meta’s board and move to an advisory role on its custom chip strategy, the companies said in a joint statement. Broadcom’s Ethernet networking technology will also be utilized to connect Meta‘s rapidly expanding clusters of AI computers.

“[The tie-up helps] build out the massive computing foundation we need to deliver personal superintelligence to billions of people,” Meta CEO Mark Zuckerberg said.

The first chip from the Meta Training and Inference Accelerator (MTIA) program, known as the MTIA 300, currently powers the social media company’s ranking and recommendation systems. Three additional chips are expected to be released by 2027. The later generations are specifically designed for inference, which is the process by which AI models respond to user queries.

In 2024, Meta announced its AI offering, designed to let users build AI-based characters as extensions of themselves. Dubbed AI Studio, the company said the offering will be available to users of Instagram, WhatsApp, and Messenger, allowing them to interact with their followers and contacts in new ways.

In order for artificial intelligence (AI) to work right within the law and thrive in the face of growing challenges, it needs to integrate an enterprise blockchain system that ensures data input quality and ownership—allowing it to keep data safe while also guaranteeing the immutability of data. Check out CoinGeek’s coverage on this emerging tech to learn more why Enterprise blockchain will be the backbone of AI.

Watch: AI is a double-edged sword