Nebius Group announced an oversubscribed $700 million private placement on December 2, 2024. The Amsterdam-headquartered, Nasdaq-listed company said it would issue 33,333,334 Class A shares at $21 each to investors including Nvidia, Accel and certain accounts managed by Orbis Investments. The proceeds were earmarked for GPU clusters, data centers, cloud platforms and developer services in the United States and other markets. Nvidia was one participant, not the disclosed sole or lead investor, and the announcement did not identify each investor’s contribution.
The transaction was equity financing rather than a loan. It gave Nebius more capital to build an AI cloud, while creating new shares for existing investors to absorb. The company did not disclose a formal post-money valuation.
The deal in brief
| Item | Details |
|---|---|
| Announcement | December 2, 2024 |
| Structure | Oversubscribed strategic equity financing/private placement |
| Gross proceeds | $700 million, as described by Nebius |
| New securities | 33,333,334 Class A shares |
| Issue price | $21 per share |
| Named participants | Accel, Nvidia and certain accounts managed by Orbis Investments |
| Placement agent | Goldman Sachs Bank Europe SE |
| Stated premium | Approximately 3% to the volume-weighted average share price since Nasdaq trading resumed |
| Planned use | GPU clusters, cloud platforms, developer tools and capacity expansion |
The share arithmetic is 33,333,334 multiplied by $21, or $700,000,014; the company rounded that figure to $700 million. Nebius’s announcement contains the transaction terms.
Why Nebius needed $700 million
AI-cloud operators must spend before customers generate revenue. A usable service requires GPUs, high-speed networking, storage, electricity, data-center space and software for scheduling and monitoring workloads. Nebius said it would combine greenfield build-to-suit data centers, colocation deployments and expansions of existing facilities.
#1 Best Overall
The company’s target customers run model training, fine-tuning, inference and other compute-intensive workloads. Nebius describes its offering as “full-stack” infrastructure spanning non-virtualized GPUs, InfiniBand networking, managed and serverless inference, MLOps tools, storage and developer services. That is broader than a marketplace that simply rents individual GPU instances.
Nebius also identified practical constraints in its risk disclosures: securing sites and power, obtaining hardware, winning customers and raising any additional capital needed for expansion. Building capacity ahead of demand creates utilization risk; waiting too long can leave a provider without the GPUs or electricity needed when customers arrive.
Who invested—and what Nvidia’s role means
The release names three investor groups: Nvidia, Accel and certain Orbis-managed accounts. It does not disclose the dollar amount invested by each. Therefore, the evidence does not support describing Nvidia as the round’s lead investor or assigning it a specific portion of the $700 million.
A strategic signal
Nvidia’s participation can reasonably be read as a vote of confidence in specialized GPU-cloud operators. Nebius was described as a preferred cloud service provider in Nvidia’s partner network and builds systems optimized for Nvidia hardware. That relationship may improve ecosystem visibility, technical coordination or access to customers, but the financing announcement does not promise preferential GPU allocation, lower prices or guaranteed supply.
The concentration question
Nvidia sells the chips that AI clouds buy and also invests in some of those operators. That alignment can accelerate deployment of Nvidia-based infrastructure, while increasing the importance of one supplier’s roadmap to participating providers. Whether it benefits customers depends on real-world capacity, pricing, reliability and competition—not on the investment alone.
Nebius’s reset as a public AI-infrastructure company
Nebius emerged after the divestment of its Russian businesses and resumed Nasdaq trading on October 21, 2024. The financing therefore marked an early capital-markets milestone for a public company rebuilding its identity around AI infrastructure, rather than a conventional venture round for a newly incorporated startup.
Accel partner Matt Weigand received board-observer rights and was expected to be nominated for election as a director at Nebius’s 2025 annual meeting. The company said its core AI-infrastructure business had around 400 engineers at the time of the announcement.
Revenue guidance and shareholder dilution
The forecast
Nebius said the financing, together with its decision not to repurchase shares, allowed it to narrow guidance to a $750 million–$1 billion annualized run-rate revenue target by the end of 2025. In this context, annualized run-rate revenue meant the final month’s revenue multiplied by 12. It was a company projection, not reported annual revenue, and should not be treated as proof that the target was achieved.
Quick wins for a faster PC:
Clear out junk files and repair common Windows errorsFree Scan →Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →The dilution
The placement created 33.33 million new Class A shares. Existing shareholders therefore owned a smaller percentage of the company after issuance, but the exact dilution percentage requires the relevant pre-transaction share count and capital structure.
Rank #4
- the perfect statement for data engineers, software developers, machine learning enthusiasts, tech nerds, and anyone whose heart genuinely beats little faster at the mention of artificial intelligence and server infrastructure.
- Lightweight, Classic fit, Double-needle sleeve and bottom hem
Nebius had previously authorized a possible repurchase of up to 81 million Class A shares at a maximum price of $10.50. After Nasdaq trading resumed and investor engagement strengthened, the board said the buyback was no longer warranted. The capital-allocation change matters: Nebius raised new equity while retaining cash that might otherwise have been used to repurchase shares.
What the financing did—and did not—prove
- It supplied capital for larger GPU and data-center deployments.
- It provided strategic validation from Nvidia and Accel.
- It did not establish that Nvidia supplied most of the money.
- It did not disclose a post-money valuation or guarantee GPU availability.
- It did not make Nebius profitable or demonstrate that its revenue forecast was achieved.
- It did not prove that Nebius offered the market’s lowest prices or best performance.
Risks investors should watch
- Capital intensity: GPUs, networking, power and facilities require large upfront spending and can become obsolete as new generations arrive.
- Utilization: Capacity built before workloads are contracted can depress returns; shortages can prevent Nebius from serving demand.
- Power and permitting: Data-center sites need electricity, interconnection capacity and regulatory approvals.
- Competition: Hyperscalers and other GPU-cloud providers can pressure prices or offer broader managed services.
- Customer concentration: Losing a large customer or failing to renew commitments can leave expensive clusters underused.
- Financing needs: Continued expansion may require further debt or equity, creating additional interest expense or dilution.
- Supplier dependence: Nebius’s systems and economics remain closely tied to Nvidia’s hardware supply and product roadmap.
What happened afterward
The December 2024 placement is historical, not Nebius’s latest financing event. Nebius later announced a September 2025 multi-billion-dollar AI-infrastructure agreement with Microsoft, documented a five-year dedicated GPU-infrastructure agreement with Meta in its 2025 Form 20-F, and announced a March 2026 partnership with Nvidia to scale its full-stack cloud across Nvidia infrastructure generations. The same filing describes an approximately $2 billion Nvidia investment in March 2026 structured around a pre-funded warrant—distinct from the December 2024 $700 million share placement.
See Nebius’s 2025 Form 20-F and the March 2026 Nvidia partnership announcement for those later developments.
Recommended Free Tools
Best Value
- Anti data center design for people concerned about AI expansion, server farm development, water usage, rural land destruction, power grid overload, corporate overreach, and the rapid growth of industrial technology infrastructure. Analog horror aesthetics!
- Perfect for supporters of local land protection, anti-AI humor, environmental awareness, small town activism, anti-corporate satire, digital dystopia art, and retro protest aesthetics. Tech apocalypse humor, server farm opposition, rural preservation!
- Lightweight, Classic fit, Double-needle sleeve and bottom hem
Who might use Nebius AI Cloud?
Nebius is aimed at AI startups, model developers, research teams and enterprises that need training, fine-tuning, batch inference or production inference capacity. Buyers can review the service at nebius.com, its documentation and console.
On August 18, 2026, the company’s pricing page listed H100 GPUs at $3.85 per GPU-hour on demand and $2.15 preemptible; H200 at $4.50 and $2.45; B200 at $7.15 and $3.95; and RTX PRO 6000 at $1.80 and $0.95. Listed L40S configurations started at $1.55–$1.82 per GPU-hour on demand, depending on CPU platform. Nebius advertised commitment discounts of up to 35% below on-demand rates for large clusters reserved for multiple months, plus free ingress and egress networking, free Kubernetes management and a $25 minimum first payment. Prices exclude applicable taxes and can change; the figures are dated to that check.
Customers should compare GPU memory, interconnects, region, storage throughput, egress terms, quota, support, security requirements and on-demand versus preemptible reliability with alternatives such as AWS EC2, Microsoft Azure, Google Cloud, CoreWeave, Lambda Cloud and RunPod. Nebius may be a poor fit for workloads requiring a specific unavailable region or managed service, guaranteed capacity based only on preemptible instances, or hyperscaler-standard procurement and certifications that have not been independently confirmed.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
The Tool Desk
Outbyte Driver Updater FREEScan for outdated or missing drivers - takes under a minuteDriver Scan →Outbyte PC Repair FREEClear out junk files and repair common Windows errorsFree Scan →
