Tesla powers up supercomputer to master AI with 10,000 Nvidia graphics cards

Tesla is about to fire up its new artificial intelligence cluster, powered by 10,000 Nvidia GPUs

Tesla powers up supercomputer to master AI with 10,000 Nvidia graphics cards
Chema Carvajal Sarabia

Chema Carvajal Sarabia

  • Updated:

If you want to master the future, you have to master Artificial Intelligence. This is true, or at least what scientists and engineers who have been working on AI for more than 30 years tell us.


Well, Tesla knows that AI alone can make its business soar tomorrow, outperforming its rivals in the automotive sector.

For that reason, Elon Musk will launch his long-awaited supercomputer on Monday. The machine will be used for various artificial intelligence (AI) applications, but the cluster is so powerful that it could also be used for demanding high-performance computing (HPC) workloads.

In fact, the Nvidia H100-based supercomputer will be one of the most powerful machines in the world.

Specifications of a computer that can dominate us all

Tesla’s new cluster will employ 10,000 Nvidia H100 compute GPUs, delivering a peak performance of 340 FP64 PFLOPS for technical computing and 39.58 INT8 ExaFLOPS for AI applications.

In fact, Tesla’s 340 FP64 PFLOPS exceeds the 304 FP64 PFLOPS offered by Leonardo, the world’s fourth highest performing supercomputer.

With its new supercomputer, Tesla is significantly improving its computing power to train its full autonomous driving technology (FSD) faster than ever before.

Not only could this make Tesla more competitive than other automakers, but it will make the company the owner of one of the world’s fastest supercomputers in AI computing.

The problem lies with Nvidia, which does not deliver

While the new H100-based cluster will dramatically improve Tesla’s training speed, Nvidia is struggling to meet the demand for these GPUs.

Therefore, Tesla is investing more than $1 billion in the development of its own supercomputer, Dojo, based on a customized and highly optimized system-on-chip.

Dojo will not only accelerate FSD training, but will also manage data processing for Tesla’s entire fleet of vehicles.

At the same time, Tesla is bringing its Nvidia H100 GPU cluster online alongside Dojo, a move that will provide the company with unprecedented computing power in the automotive industry.

Elon Musk recently revealed that Tesla plans to spend more than $2 billion on AI training in 2023 and another $2 billion in 2024 specifically on FSD training computing.


Some of the links added in the article are part of affiliate campaigns and may represent benefits for Softonic.

Chema Carvajal Sarabia

Chema Carvajal Sarabia

Journalist specialized in technology, entertainment and video games. Writing about what I'm passionate about (gadgets, games and movies) allows me to stay sane and wake up with a smile on my face when the alarm clock goes off. PS: this is not true 100% of the time.

Latest from Chema Carvajal Sarabia