Nebius Group (NBIS): Microsoft Deal and Updated Valuation Model
As I anticipated several times, Nebius has finally signed its first contract with a Big Tech company.
In my list of catalysts from three months ago, I highlighted the New Jersey data center as a critical one, specifically because it would enable Nebius to land a transformational deal with a major tech corporation, whether a hyperscaler or a frontier AI lab.
That moment has arrived. Nebius just announced a $17.4–19.4B contract over five years with Microsoft to deliver dedicated capacity from its new Vineland, New Jersey data center, starting later this year.
Let’s dive into it.
Why This Deal Matters
After the announcement, NBIS stock jumped more than 50%.
But why is this deal so important?
To start, the contract’s value exceeds the company’s entire market cap. To be honest, while I expected a large deal tied to this data center, the size is about five times larger than I had envisioned.
Here’s what Arkady Volozh, Founder and CEO of Nebius, said in the press release:
“I’m happy to announce the first of these contracts, and I believe there are more to come.”
And during the last earnings call, he stated:
“We have nearly closed on two substantial new greenfield sites in the U.S. Each one will be able to deliver hundreds of MWs of power in 2026, and we’re confident we’ll be making an announcement about that soon.”
This means Nebius is preparing at least two more data centers of similar scale by 2026, each capable of supporting large deals like the Microsoft one.
Connecting the dots, it seems reasonable to expect at least one more major contract by next year, potentially more.
As I explained in my article back in June, these deals matter not only for their size but also for what they represent. When CoreWeave began securing long-term agreements with blue-chip AI players, its valuation soared, and the same dynamic is now playing out with Nebius.
Large-scale contracts typically bring:
Revenue visibility over multiple years
Utilization guarantees that reduce infrastructure risk
External validation that Nebius’ tech and operations meet enterprise-grade standards
The key takeaway: this deal significantly de-risks Nebius’ story.
I’ve been asked: “Isn’t Nebius supposed to pursue deals that include its value-added services, instead of just bare metal, which has thinner margins and less differentiation?”
Here’s the nuance:
Bare metal isn’t commoditized yet. The supply-demand imbalance is still massive. For example, CoreWeave earns 70%+ gross margins on such deals, despite not having Nebius’ vertical integration. I expect strong margins from this contract with Microsoft.
Nebius’ long-term vision remains focused on enterprises, outcomes, and differentiated services. But the company has always intended to serve the full spectrum: AI-native startups, hyperscalers, frontier labs, enterprises, and even national AI initiatives.
If/when bare metal ever does become commoditized, it would likely mean AI adoption is already widespread across enterprises, creating ample demand for Nebius’ AI cloud and higher-value services to absorb capacity once contracts like Microsoft’s expire (in case Microsoft doesn’t want to extend it).
Securing a $17B contract provides much more than revenue. It validates Nebius’ technology, enhances creditworthiness through utilization guarantees, and creates a stronger position for fundraising, all of which are essential when billions of dollars are required to keep scaling aggressively.
In short, even if it’s “just bare metal”, this deal opens the door for Nebius to further invest in its product offerings, as Roman Chernin recently emphasized in his CNBC interview. You can’t dismiss a deal worth more than your market cap, especially when the counterparty is Microsoft.
Some people have asked: “Why would Nebius land a Big Tech deal before companies like CIFR or IREN, which have far more gigawatts of capacity?”
The answer lies in understanding active, connected, and contracted power:
Active power = GPUs already installed and running; capacity that is fully operational today.
Connected power = active power plus fully provisioned capacity that can go live immediately once GPUs are installed.
Contracted power = capacity secured through agreements but still requiring construction, substation upgrades, or grid connections before it can become “connected.”
CIFR and IREN may show larger contracted capacity, but turning that into connected power, and eventually into active power, takes significant time and investment.
And importantly, AI infrastructure is about more than just GPUs and gigawatts. Even for bare metal, reliability, efficiency, hardware quality, security, scalability, and performance all matter, especially for a hyperscaler like Microsoft, which will pay more for guaranteed quality. While Nebius is a young company, its team of hundreds of engineers has decades of experience working together and at scale, having built over 20 data centers at Yandex. That track record probably gives Microsoft confidence in Nebius’ ability to execute.
This doesn’t mean retrofitted Bitcoin miners like CIFR and IREN won’t eventually land big contracts. Given the sector’s massive supply-demand imbalance, they likely will IMO. But it’s clear why a hyperscaler would select Nebius as a preferred partner.
Financing Update
In the days following the Microsoft announcement, Nebius provided a financing update that highlights just how much the deal has strengthened its creditworthiness.
1) Convertible Senior Notes
• $1.375B of 1.00% notes due 2030
• $1.375B of 2.75% notes due 2032
• Initial conversion price: $138.75/share
• Effective conversion price at maturity: $159.56/share
• Upsized from $2.0B → demand was strong
• Repayable in cash, shares, or a mix
2) Class A Share Offering
• $1.0B offering at $92.50/share (~4.5% dilution)
• 30-day underwriter option for an extra $150M
I’m very pleased with these terms.
Just three months ago, Nebius raised $1B in convertibles at ~2.5% interest with an effective conversion price of ~$63. Now, it has secured almost $4B at a lower average coupon of 1.875% and a conversion price 153% higher than before. That’s a dramatic improvement in just a few months, driven largely by the Microsoft deal.
The structure is smart: low coupons, high conversion premiums, strong investor demand. Arkady clearly knows what he’s doing.
I’ve always said dilution was part of the story, and based on these terms, I’m very comfortable. The capital will fund both capacity expansion and product expansion/improvements beyond Microsoft’s deal. Unlike CoreWeave, which is heavily hyperscaler-focused, Nebius emphasizes building its broader AI cloud (“AWS of AI”) while treating hyperscaler deals as additional fuel for acceleration.
This financing round is especially important because it gives Nebius the flexibility to wait for better valuations before monetizing non-core business units for non-dilutive capital.
More Institutional Coverage to Come
Unlike the previous raise, where Goldman Sachs acted as sole placement agent, the recent Class A share offering also included Morgan Stanley, BofA Securities, and Citigroup as book-running managers. I wouldn’t be surprised to see more institutional coverage follow, something I’ve long flagged as a catalyst.
Nebius has been underfollowed ever since going public, partly because of its unusual listing process, which didn’t involve a traditional IPO roadshow. While it was gradually gaining attention, this Microsoft deal changes everything. You can’t ignore a $17.4B contract with Microsoft, and I believe this will trigger a wave of new institutional eyes on the stock.
Valuation Model Update
I updated my Valuation Model on Nebius less than a month before the Microsoft deal. You can read the full details here:
Here were my final results (Price Target for YE 2025):
I don’t see any reason to adjust my Non-Core Business Units valuation since nothing has changed there. That said, I still think my assumptions may prove conservative. Check out all the details in my full Valuation Model.
For the core business, however, the Microsoft deal forces an update.
Previously, I assumed Nebius would reach 1 GW of active power by the end of 2027, given how challenging it is to secure enough GPUs and funding to fully activate the >1 GW of contracted power guided for the end of 2026. Now, after the deal and financing update, I’m moving my estimate forward by one year.
Still, I’m applying a 15% margin of safety to account for potential supply chain issues and data center construction delays (like with New Jersey, though it still met Nebius’ internal timeline). That gives me an estimate of 850 MW of active power by YE 2026.
I’m also raising my EBIT multiple range from 15–25x to 20–30x. Why? Because predictability of revenue and cash flows is, in my opinion, one of the biggest drivers of valuation multiples. With this $17.4B contract, 5x larger than what I expected, Nebius now enjoys far greater visibility, utilization guarantees, and external validation.
For context, CoreWeave still trades at ~20x 2026 EBIT estimates, even after its stock dropped more than 40% from all-time highs (1.25 years forward multiples, while I’m applying them as TTM).
I could justify lowering the discount rate, since the business is de-risked. But I’ll keep it at 15% given the ongoing uncertainty around long-term unit economics in the sector. Also, as I said, I was already expecting a large deal to happen at any time, so my choice of using 15% had that in mind.
On dilution: my prior model already assumed 10% additional dilution to account for future raises. So far, the share offering was ~4.5%, and the convertibles will eventually increase that if converted (which I’d welcome, since it would mean NBIS is trading much higher).
Everything else in the model remains the same. For full context, I recommend revisiting my original assumptions.
Updated Valuation Scenarios (Core Business Only)
• Bear Case (20x 2027E EBIT, 15% discount rate): $20.6B
• Base Case (25x 2027E EBIT, 15% discount rate): $25.7B
• Bull Case (30x 2027E EBIT, 15% discount rate): $30.9B
Here’s my updated Valuation Model Post-Microsoft deal (YE 2025):
Final Thoughts
When I first started buying NBIS in January and later made it my largest position (and largest investment ever), I did it exactly because I saw it as an asymmetric opportunity with low risk and significant upside potential. I knew it would keep me comfortable owning the stock even if it doubled or tripled, and that’s exactly what happened.
This “target” of $116.72/share is, hopefully, just the beginning of something much larger. I believe this contract will mark the beginning of large institutional inflows, similar to what happened with other companies that were first discovered by retail investors (for example, my investment in RKLB, which has already more than 10x).
I have already had to update my Valuation Model three times this year, and that speaks volumes about the execution of this team. If they continue to work like they have been, I will likely have to update it again soon, especially as more large contracts are signed.
Moreover, I have a feeling that if this bull market continues, ClickHouse will take advantage and IPO in 2026, probably at a huge valuation, as it is one of the hottest AI startups out there. The same goes for Avride and Toloka, which may end up having much larger importance to NBIS’ valuation than what I accounted for.
I don’t want to be overly optimistic, and that’s why I kept most of my assumptions somewhat conservative (as always, this is highly subjective), but the fact that there are so many variables that could work out much better than I am currently factoring in makes me believe that the upside is simply too large for me to feel the urge to trim my position. My time horizon for this investment has always been at least a few years, and although that could obviously change as market conditions and the company’s fundamentals evolve, it surely remains much longer than just a few months.
I truly believe this is just the beginning of Nebius’ story, so I am not likely to trim my position at current levels.
That’s it. Thanks for reading!
Note: I’m admittedly biased, so please make sure to study the company yourself and understand all the dynamics behind this valuation, as well as the broader risks associated with this investment.
Don’t forget to follow me on X (@mvcinvesting), where I share daily news about the company.
Disclaimer: As of this writing, M. V. Cunha holds a position in Nebius Group (NBIS) at $25.67/share.
Thanks for the great and clear, yet thorough, update as always. much appreciated!
Thank you sir!