Tech
Nebius data center
(Photo: Markus Pentikäinen)

How Nebius emerged from Russia and became one of Nvidia’s top-performing investments

The small Amsterdam-based AI data center startup rose from the ashes of a Russian tech giant.

Nvidia’s investment in AI cloud computing startup (and customer) CoreWeave is turning out to be quite a smart move for the company. CoreWeave’s shares have been on a tear since its IPO in March — almost 300%. Nvidia, whose GPUs fill CoreWeave’s data centers, stepped in to shore up that IPO and now owns about 7% of the company, a stake worth $3.7 billion

The CoreWeave investment is Nvidia’s largest disclosed holding by far (excluding investments by NVentures, Nvidia’s venture capital arm), but it isn’t the only data center business Nvidia’s invested in that’s showing big gains.

From the ashes of Yandex

The next best performer in Nvidia’s holdings is AI cloud computing company Nebius. Since the end of Q1, the value of Nvidia’s Nebius holdings has risen about 140%. While the $60 million stake is much, much smaller than Nvidia’s stake in CoreWeave, Nebius is not like other startups. The company has an interesting history that’s led it to become a rising competitor in building out AI computing infrastructure. 

Nebius emerged from the Russian tech giant Yandex. 

Founded in 1997, Yandex — which essentially became Russia’s Google, Uber, Spotify, and PayPal all rolled into one — made a splash back in 2011 with its $1.3 billion IPO, which at the time was the largest tech IPO since Google’s in 2004. Fast-forward to 2022, when Russia’s invasion of Ukraine led to the company’s stock being suspended from trading on Nasdaq due to EU sanctions. 

Russian President Vladimir Putin Visits Yandex In Moscow
Russian President Vladimir Putin listens to Arkady Volozh while visiting Yandex on September 21, 2017 in Moscow, Russia (Mikhail Svetlov/Getty Images)

Nebius’ founder and CEO is Arkady Volozh. Based in Tel Aviv, Israel, the executive was also the cofounder and CEO of Yandex. Volozh found himself subject to EU sanctions after the war began, but they were lifted in March 2024 after Volozh moved hundreds of engineers out of Russia and publicly disavowed the Russian invasion, calling it “barbaric” but also saying, “Although I moved to Israel in 2014, I have to take my share of responsibility for the country’s actions.” 

Last July, Yandex NV (based in Amsterdam) officially sold off its Russian holdings to a group of Russian investors for $5.4 billion (to continue operating as Yandex inside Russia), the largest corporate exit from Russia since the war began. The remaining business renamed itself Nebius Group. In October, the Nasdaq allowed trading of the rebranded company. 

The company sought to rebuild itself around the four businesses in the Nebius Group holding company: 

  • Nebius, the core business with a focus on AI cloud computing infrastructure, powered by “hundreds of thousands of GPUs

  • AVRide, an autonomous vehicle and robotics company

  • Toloka AI, a data training company backed by Shopify CTO Mikhail Pakakhin and Amazon founder Jeff Bezos’ Bezos Expeditions

  • TripleTen, an educational technology company that helps people develop the skills to work in the technology sector 

In an SEC filing this February, Nvidia disclosed its investment in Nebius, worth about $33 million by the end of Q4 2024, but is up 140% since Q1 2025. Like fellow Nvidia investee CoreWeave, the close relationship with the chip giant lets Nebius offer early access to the latest and greatest GPUs.

Sovereign AI

Nebius is looking to position itself to serve the growing market for sovereign AI, especially in Europe, which is seeing a flurry of AI infrastructure activity. Last week, Nvidia CEO Jensen Huang was in Paris, where he highlighted Nebius as one of Nvidia’s partners building AI cloud infrastructure in Europe. Last week, Nebius also announced its entrance to the UK market and that it would be among the first to offer Nvidia’s latest Blackwell GPUs in Europe.

In May, the Israel Innovation Authority announced that Nebius had been selected to build and run a “16,000 petaflop” national AI supercomputer for the country, utilizing 4,000 Nvidia Blackwell GPUs. Volozh is an Israeli citizen, and moved to the country in 2014 after Russia first invaded Crimea. The supercomputer will be used to develop “national AI models” and for research.  

Nebius currently has a market cap of $12 billion, and revenue for Q1 was up 385% year over year to $55.3 million, with adjusted net losses of $92.5 million. The company spent $544 million on capital expenditures in Q1 as it continues to invest in pricey GPUs for its AI data center business. Today, Nebius has data centers in Iceland, Finland, France, Missouri, and a new facility under construction in New Jersey.

In a Q1 2025 letter to investors, Volozh wrote: 

“Delivering on our rapid growth plans will continue to require considerable investment. In this respect, we have been in a uniquely favorable position to fund our growth since we formally launched Nebius last July. We started with a significant amount of capital — $2.5 billion from our divestment — and raised an additional $700 million from top tier investors in December, giving us more than $3 billion in cash. This has enabled us to quickly build and scale our AI cloud business.”

It remains to be seen if Nebius will follow the same meteoric rise as CoreWeave. But the close partnership with Nvidia gives Nebius a similar advantage to get early access to Nvidia’s latest and most powerful GPUs

Correction (June 17 2025): In an earlier version of this article, we said Nebius was Nvidia’s second-biggest investment when it is Nvidia’s second-fastest-growing investment.

More Tech

See all Tech
tech

Tesla’s made-in-China vehicle sales jumped 36% in April

Tesla’s sales of made-in-China vehicles — sold across China, Europe, and other international markets — rose 36% year over year to 79,478 units in April. The increase marks the sixth straight month of annual growth in sales of vehicles made in the worlds largest manufacturing economy, suggesting the EV maker’s overseas business may be stabilizing after a difficult stretch.

That said, China wholesale deliveries fell from March, even as overall new energy vehicle sales rose 7% during the period.

Later this month, the China Passenger Car Association will report China-only sales, offering a clearer picture of performance in Tesla’s second-largest market.

Later this month, the China Passenger Car Association will report China-only sales, offering a clearer picture of performance in Tesla’s second-largest market.

tech

Anthropic’s scramble for compute now includes rival xAI

Another day, another major partnership with an AI rival. This time, Anthropic signed a deal with SpaceX’s xAI to access compute from its Colossus 1 data center to help it improve capacity for its Claude Pro and Claude Max subscribers. Just yesterday, The Information reported that Anthropic planned to spend $200 billion on Google Cloud services over the next five years. As Sherwood News’ Luke Kawa wrote:

“Anthropic has been a victim of its own success: the popularity of Claude Code and Cowork have revealed compute constraints and left users frustrated by caps. In response, the Claude developer has embarked upon a mad scramble for compute, striking or expanding deals with CoreWeave, Amazon, Google, and Broadcom.”

Now, it’s adding xAI to the list — even as the Elon Musk company builds a competing model.

In less terrestrial news, xAI said that as part of the agreement, Anthropic “expressed interest in partnering to develop multiple gigawatts of orbital AI compute capacity.”

“Anthropic has been a victim of its own success: the popularity of Claude Code and Cowork have revealed compute constraints and left users frustrated by caps. In response, the Claude developer has embarked upon a mad scramble for compute, striking or expanding deals with CoreWeave, Amazon, Google, and Broadcom.”

Now, it’s adding xAI to the list — even as the Elon Musk company builds a competing model.

In less terrestrial news, xAI said that as part of the agreement, Anthropic “expressed interest in partnering to develop multiple gigawatts of orbital AI compute capacity.”

tech

SpaceX and Tesla’s Terafab could cost $119 billion — far more than expected

The initial phase of SpaceX and Tesla’s joint chip production effort, called Terafab, could cost $55 billion, with additional phases adding up to $119 billion in capital investment, Reuters reports, citing a notice posted on a Texas county website. Ultimately the goal of Terafab is to build enough in-house AI chip capacity to supply both companies.

The price tag is also higher than expected. Morgan Stanley had previously estimated Terafab would cost $34 billion to $45 billion.

Fortunately for Tesla, whose capex is expected to skyrocket this year, much of the early spending will sit on SpaceX’s balance sheet.

Here’s Musk on the last earnings call:

“SpaceX is going to take care of like the initial phase of the scaled up Terafab... Any kind of intercompany thing has to be approved by both the SpaceX and Tesla board of directors. It’s got to go through a conflict resolution. It’s going to have, unfortunately, a lot of complexity because we’ve got to make sure Tesla shareholders are served and SpaceX shareholders are served, and strike the right balance there.”

The price tag is also higher than expected. Morgan Stanley had previously estimated Terafab would cost $34 billion to $45 billion.

Fortunately for Tesla, whose capex is expected to skyrocket this year, much of the early spending will sit on SpaceX’s balance sheet.

Here’s Musk on the last earnings call:

“SpaceX is going to take care of like the initial phase of the scaled up Terafab... Any kind of intercompany thing has to be approved by both the SpaceX and Tesla board of directors. It’s got to go through a conflict resolution. It’s going to have, unfortunately, a lot of complexity because we’ve got to make sure Tesla shareholders are served and SpaceX shareholders are served, and strike the right balance there.”

tech

Apple to let users choose between Anthropic, Google, and OpenAI models

Apple has been inching toward letting outside AI power its devices — and now it’s going further.

The company plans to let users choose between rival AI models across iOS 27, due this fall, expanding beyond ChatGPT to include players like Google and Anthropic, Bloomberg reports. The difference this time: deeper integration, with outside models powering features like Siri, writing tools, and image generation across the system.

Currently, Apple’s voice assistant, Siri, gives users the ability to query ChatGPT, but doing so requires a clunky extra step and usage has been poor. Meanwhile, Apple’s own AI tools have fallen short. (Apple has decided to use Google’s Gemini to power Siri in the future.) It’s not clear users care which AI is under the hood — as long as it works.

Currently, Apple’s voice assistant, Siri, gives users the ability to query ChatGPT, but doing so requires a clunky extra step and usage has been poor. Meanwhile, Apple’s own AI tools have fallen short. (Apple has decided to use Google’s Gemini to power Siri in the future.) It’s not clear users care which AI is under the hood — as long as it works.

Latest Stories

Sherwood Media, LLC produces fresh and unique perspectives on topical financial news and is a fully owned subsidiary of Robinhood Markets, Inc., and any views expressed here do not necessarily reflect the views of any other Robinhood affiliate, including Robinhood Markets, Inc., Robinhood Financial LLC, Robinhood Securities, LLC, Robinhood Crypto, LLC, Robinhood Derivatives, LLC, or Robinhood Money, LLC. Futures and event contracts are offered through Robinhood Derivatives, LLC.