Nebius Group N.V.

Nebius Group has acquired inference optimization startup Inferize and signed new long-term AI data center capacity deals, moves that aim to boost GPU utilization on its Token Factory platform and deepen its role in global AI infrastructure. The company's recent inclusion in the FTSE All-World Index came alongside fresh analyst coverage and large customer commitments from partners including Meta Platforms and Palantir. Nebius Group's narrative projects $29.7 billion in revenue and $2.1 billion in earnings by 2029, requiring 179.9% yearly revenue growth and about a $2.1 billion increase in earnings from $42.4 million today. Some of the most optimistic analysts were already modeling Nebius to reach about US$48.5 billion in revenue and US$1.1 billion in earnings by 2029, well above consensus. The Inferize deal directly targets Nebius's cost base and unit economics by cutting the idle GPU tax on Token Factory, though financing needs and potential dilution remain key risks.

Nebius has acquired Inferize, an Israel-based AI startup specializing in reducing idle GPU capacity and accelerating the deployment of large models, the neocloud announced today. Inferize's technology targets so-called cold starts, the time models need to load before serving requests, which can leave GPUs idle during a launch as demand spikes and new instances occur. No financial details were disclosed, but Calcalist estimated the deal ranged from $100M to $150M. Nebius plans to add Inferize's experienced engineers to its inference team, with Chief Technology Officer Danila Shtan saying the team's contribution will extend well beyond this first integration. The acquisition follows Nebius's recent purchases of Eigen AI in a deal valued at $643M and Clarifai, whose value was not disclosed, to improve inference and model optimization.

Palantir Technologies reported continued expansion of its AIP-based commercial AI business across healthcare, defense, and manufacturing, raised full-year revenue guidance on robust demand, and named Nebius its preferred sovereign AI infrastructure partner, integrating Nebius compute into AIP. The company's narrative projects $23.0 billion revenue and $10.2 billion earnings by 2029, requiring 55.1% yearly revenue growth and about a $7.2 billion earnings increase from $3.0 billion today. That forecast yields a $191.68 fair value, in line with its current price. Some of the lowest estimate analysts assumed revenue of about US$17.7 billion and earnings of roughly US$6.3 billion by 2029, yet still saw downside risk. Investors continue to weigh valuation risks, competitive intensity, and concentration of growth in AI-related contracts.

Nebius Group will raise its Nvidia GPU rates by roughly 17 to 21 percent and its AMD EPYC Genoa CPU service rates by 25 percent beginning October 1, citing continued demand for computing capacity and limited computing resources. The price increase, which the company said should translate into higher revenue beginning in the fourth quarter, came alongside a BNP Paribas price target hike of 53 percent to $399 from $260, with the rating raised to outperform from neutral. Nebius shares rose 7.44 percent on Thursday to close at $243.48 apiece. In the second quarter, the company swung to a net loss of $190.4 million from a $502.5 million net income a year earlier, while adjusted net loss narrowed by 64 percent to $33.2 million from $91.5 million, and revenues jumped by 454 percent to $582.3 million from $105.1 million. Hedge fund holdings in the stock more than tripled to $8.28 billion from $2.36 billion quarter-on-quarter, with 86 funds holding positions, up from 60 in the first quarter.

Palantir Technologies and Nebius Group announced a strategic partnership on September 8 to bring Nebius's AI-native compute infrastructure into Palantir's commercial platform. Palantir designated Nebius as its preferred sovereign AI infrastructure partner, integrating Nebius compute and inference endpoints directly inside the Palantir enterprise perimeter. The alliance allows commercial clients to deploy open AI models and run compute-heavy workloads while maintaining strict control over proprietary data, and the companies will also collaborate on modular data center deployments to rapidly scale compute capacity where power is available. For Palantir, the deal provides dedicated high-performance compute capacity supporting its rapid U.S. commercial expansion and 157% net dollar retention rate, while for Nebius it validates its full-stack AI cloud infrastructure and unlocks direct access to Palantir's enterprise customer base. The partnership carries execution and capital risks for both sides, with Palantir facing higher hosting and delivery expenses and Nebius remaining subject to persistent EBIT losses and negative free cash flow as it aggressively builds out data centers and acquires GPUs.

Palantir Technologies shares climbed more than 3% Wednesday and moved above $190, their highest level in nearly a year, as new government contracts and expanding AI deployments added to the stock's momentum. The latest boost came from the U.S. Army, which awarded Palantir a $48.1 million contract to build an enterprise ammunition-management system that will replace nine older systems with a single platform. The stock has gained 69% over the past three months, helped by stronger-than-expected results and growing adoption of its Artificial Intelligence Platform. Palantir is also expanding overseas, with CEO Alex Karp meeting the presidents of Poland and Lithuania this week to discuss potential technology investments, and the company announcing partnerships with Nvidia, Nebius and Method Security, plus a separate deal with Fujitsu aimed at expanding its software across enterprise networks in Japan. The FAA recently began using Palantir's AI-powered SMART system at three Washington-area airports to identify congestion and other problems before they develop, and Rosenblatt analyst John McPeake reiterated a Buy rating and a $225 price target, citing additional FAA opportunities.

Nebius Group reported second-quarter revenue of $582.3 million, up 454.04% year over year and ahead of the $574.65 million consensus, with its AI Cloud segment alone contributing $574.9 million, up 514%. The company's remaining performance obligations stand at $37.5 billion, dwarfing current annual revenue, while deferred revenue climbed to $5,975.2 million from $1,577.5 million at year-end. Meta signed a $27 billion five-year agreement with Nebius covering $12 billion of compute plus $15 billion of flexible capacity, and NVIDIA holds a $2 billion strategic equity stake plus pre-funded warrants. Management reaffirmed full-year 2026 revenue guidance of $3.0 billion to $3.4 billion and annualized run-rate revenue of $7 billion to $9 billion by year-end at roughly a 40% adjusted EBITDA margin, after Q2 group adjusted EBITDA printed at $236 million, a 41% margin. CEO Arkady Volozh said he could sell out 2026 capacity today but is holding out for better terms, with shares up 182.09% year to date.

On September 22, major A-share indices opened collectively higher, with AI chips, CPO, semiconductors, components, and media sectors leading gains. Huaci Shares hit the daily limit for a sixth straight session. The company said its Vietnam production base phase-one project is still in the capacity ramp-up stage, with limited short-term capacity that is unlikely to offset the impact of overseas trade barriers, and it expects third-quarter earnings to continue declining year on year. Two new stocks debuted that day. Zhongsu shares listed on the Shenzhen Stock Exchange's ChiNext board, opening up 384.37 percent. As of press time, the stock was up more than 400 percent at 279 yuan per share, versus an issue price of 55.28 yuan per share, meaning a single lot of 500 shares gained more than 110,000 yuan. On the Beijing Stock Exchange, Century Digital opened higher and was up 314.81 percent as of press time at 65.04 yuan per share, with a total market value of about 2.9 billion yuan, versus an issue price of 15.67 yuan per share. In news, the popularity of Meta's AI agent Muse boosted expectations for server CPU demand. ARM, Intel and several other semiconductor chip stocks surged more than 10 percent, and AMD's market value surpassed 1 trillion dollars for the first time. Computing power leasing provider Nebius announced it is raising on-demand GPU cloud service prices by an average of about 20 percent.

Nebius Group shares jumped nearly 9% in Thursday premarket trading after the AI cloud provider reportedly raised prices across several compute services, a signal that demand for scarce GPU capacity remains strong. According to a customer communication circulated online, Nebius plans to increase prices for several on-demand GPU offerings beginning October 1, covering Nvidia's H100, H200, B200 and B300 accelerators. The reported increases extend beyond GPUs: prices for AMD EPYC Genoa CPU compute are expected to rise 25% to $0.015 per vCPU-hour from $0.012, while memory pricing tied to Genoa systems is reportedly increasing about 41% to $0.0045 per GiB-hour from $0.0032. Nebius has not publicly confirmed the pricing changes and did not immediately respond to a request for comment. The market reaction spread beyond Nebius, with IREN shares rising about 5.1% in premarket trading and CoreWeave gaining roughly 6.1%, suggesting investors read the reported increases as a positive signal for the wider neocloud industry.

Nebius Group notified customers of higher on-demand rental rates for a range of AI GPU capacity effective October 1, a rare same-week read-through on pricing power in AI compute that split the neocloud trade three ways. Nebius stock climbed 10% to $229.67 as the party setting the new price card, while IREN Limited rose 6% to $45.20 on the read-through that scarce GPU capacity can be re-rented at higher hourly rates. CoreWeave slipped 2% to $82.10, with the divergence attributed to concerns over how much of its capacity is already committed at older rates, how much debt sits against that capacity, and how fast free cash flow will turn. The Invesco QQQ Trust was up 1.62% to $716.12, while the Global X Data Center & Digital Infrastructure ETF trailed at up 1% to $27.60, a lag suggesting the diversified digital infrastructure basket is not capturing the neocloud repricing on its own. JPMorgan double-upgraded IREN to Overweight from Underweight on September 14, a call that predates this move by three sessions and serves as context rather than catalyst.

Nebius Group shares jumped over 6% in extended trading on September 16 after the company revised pricing strategies across its on-demand AI compute resources. The move comes even as the stock remains down roughly 22% over the trailing month amid balance sheet stress, equity dilution fears, and sector rotation. Operational results show significant traction, with quarterly revenue surging 454% year-over-year to $582.3 million in the latest Q2 2026 figures and trailing twelve-month revenue reaching $1.36 billion. Powering that expansion is a contracted backlog approaching $40 billion, anchored by a $27 billion five-year deal with Meta Platforms and a $17.4 billion contract with Microsoft, alongside a $2 billion equity investment from Nvidia. Management also raised year-end 2026 contracted power capacity guidance from over 4 GW to over 5 GW, though short interest sits at 17% of the public float as investors weigh execution risk on an aggressive $20 billion to $25 billion CapEx plan for 2026.

Generac Holdings surged 33.7% in premarket trading after announcing a long-term agreement with Amazon to supply industrial backup generators for the technology giant's global data centers. The agreement calls for initial deliveries worth about $2.4 billion across 2027 and 2028, while the total value of the arrangement could reach as much as $8 billion. Amazon will also receive a warrant to purchase about 1.69 million Generac shares at an exercise price of approximately $200.93 per share, representing nearly 3% of the company's shares outstanding. Nebius Group NV rose 11.1% in pre-open trading after notifying customers of broad-based price increases across its on-demand GPU cloud services effective Oct. 1, with prices rising approximately 17% for H100 instances, roughly 21% for the latest Nvidia B300 GPU, and 25% for AMD EPYC Genoa CPU rates. BCB Bancorp fell 4.9% in premarket trading after pricing an underwritten public offering of 11 million common shares for gross proceeds of $85.25 million and disclosing it is marketing approximately $210 million of problem loans, while expecting a net loss of between $126.2 million and $136.1 million for the third quarter of 2026.

Shares of Nebius Group N.V., or NBIS, rose 6.58% in overnight trading to $223.14 after reports that Nebius plans to raise its on-demand GPU rental rates effective October 1, 2026, the second price increase this year. Under the new rates, the H100 GPU goes from $3.85 to $4.50 per GPU-hour, a rise of about 17%. The H200 goes from $4.50 to $5.40, up 20%. The B200 goes from $7.15 to $8.50, up about 19%. And the B300 goes from $7.85 to $9.50, up roughly 21%. Yuanta Securities Thailand said the simultaneous increases across both older GPUs and newer Blackwell-family chips reflect that demand for artificial intelligence computing power remains high and that Nebius is gaining more pricing power. The company also sees forward demand extending beyond 24 months, with some customers already booking capacity through 2028. If system utilization stays high, an average price increase of about 20% would significantly boost revenue per GPU and raise returns on data center investment. Among depositary receipts tied to NBIS on the Thai market, NBIS01 was unchanged at 4.12 baht with 2.91 million baht in trading value. NBIS03 rose 0.05 baht, or 0.71%, to 7.05 baht with 4.03 million baht in trading value. NBIS23 was unchanged at 7.05 baht with 797,000 baht in trading value. And NBIS80 was unchanged at 7.00 baht with 8.22 million baht in trading value.

Palantir Technologies and Nebius Group announced a partnership on 8 September 2026 making Nebius Palantir's preferred sovereign AI infrastructure provider, integrating Nebius' AI-native cloud and inference endpoints into Palantir's commercial perimeter to give eligible customers more control over compute, data, and models. The alliance positions Nebius as an AI-first cloud platform built for demanding AI workloads and could broaden its reach through Palantir's enterprise distribution and modular data-center deployments. The news comes alongside Nebius' first senior secured debt facility of about US$775.0 million, backed by GPU assets and contracted cash flows, underscoring how the company is funding and filling significant new capacity. Nebius Group's narrative projects $30.3 billion revenue and $2.4 billion earnings by 2029, with a $312.67 fair value implying 49% upside to its current price, while some of the lowest ranked analysts had assumed heavy spending would keep Nebius unprofitable even with revenue reaching about US$19.1 billion by 2029. The Palantir deal strengthens Nebius' case as a sovereign AI provider and could reinforce near-term demand visibility, but it does not remove the key near-term risk around heavy capital spending and the company's ability to translate rapid revenue growth into sustainable margins.

Nebius Group shares have dropped 23.58% over the past month, sliding from $277.68 on August 14 to $212.19, even as Arete Research reiterated a Buy rating and raised its price target to $415 from $380, implying roughly 96% upside. The pullback followed a blistering run that still leaves the stock up 153.5% year to date, and it came alongside a broad neocloud selloff in which CoreWeave fell 21.17% to $82.98 and Applied Digital dropped 21.22% to $24.58. Nebius reported Q2 2026 revenue of $582.3 million, up 454% year over year and ahead of the $574.65 million consensus, with GAAP EPS of -$0.68 beating the -$0.8633 estimate, while its AI Cloud segment posted a 50% adjusted EBITDA margin. The bull case rests on $37.5 billion in remaining performance obligations, which management described as $40 billion in contracted backlog on the Q2 call, supported by multi-year hyperscaler agreements with Microsoft and Meta and scaling past 1 GW of connected power. Management reiterated 2026 guidance of $3.0 billion to $3.4 billion in revenue, $7 billion to $9 billion in year-end ARR, and roughly 40% adjusted EBITDA margin, though investors have focused on 12.7 million ATM shares issued at an average $223.60, an $8.13 billion H1 capex bill, and customer concentration with three customers at 24%, 21%, and 14% of Q2 revenue.

Nebius Group ended the second quarter of 2026 with roughly $40 billion in total contracted backlog and customer commitments, including $37.49 billion in remaining performance obligations, against a market capitalization near $49.49 billion. About 70% of second-quarter deals carried upfront prepayments, and customer prepayments are expected to provide more than $9 billion of funding in 2026 alone, covering 50% to 60% of the associated capital expenditures. Group revenue for the quarter reached $582.30 million, up 454.04% year over year and ahead of consensus, while the Nebius AI Cloud segment grew 514% and posted a 50% adjusted EBITDA margin; group adjusted EBITDA was $236 million versus a prior-year loss of $21 million, and annualized run-rate revenue hit $3 billion in June, up from $1.9 billion at the end of March. A capacity auction cleared 15% above the highest price Nebius had ever charged for Blackwell-generation chips and 20% above its Blackwell pipeline pricing, with short-duration premium contracts at $40 million to $50 million per megawatt versus $20 million to $25 million per megawatt on core deals. Microsoft signed a Nebius agreement valued at $17.4 billion to $19.4 billion and NVIDIA injected $2 billion of strategic equity, while three customers accounted for 24%, 21% and 14% of second-quarter revenue and convertible debt carried a carrying value of $8.5 billion against a fair value of $20.8 billion. Contracted power is set to step up to 5 gigawatts by year-end 2026, and CEO Arkady Volozh said on the second-quarter call that Nebius could sell its entire 2027 capacity today on those terms.

Oracle's blowout AI cloud results lifted shares of fellow AI infrastructure names CoreWeave, Nebius Group and IREN in premarket trade Friday. CoreWeave climbed around 2.5%, Nebius advanced approximately 2.4% and IREN gained about 1.4% as investors reacted to Oracle's disclosure that cloud infrastructure sales rose 121% year-over-year to $7.4 billion and that the business signed more than $30 billion in new cloud contracts connected to AI. The read-through matters because CoreWeave, Nebius and IREN are all investing heavily in Nvidia GPUs, electricity and data centers on the thesis that clients need ever more computing capacity. Recent company updates have added momentum: CoreWeave announced a field engineering solution to help companies deploy AI models, Nebius partnered with Palantir, and IREN moved ahead with its planned 2-gigawatt Texas data-center project. Wall Street still sees substantial upside, with average analyst estimates from TipRanks pointing to roughly 73% for IREN, 56% for CoreWeave and 26% for Nebius, though the real test remains whether returns on capital can outpace the huge cost of building the infrastructure.

NVIDIA Chief Executive Officer Jensen Huang said generative AI demand and the end of Moore's law are driving a surge in computing infrastructure spending that could reach $3 trillion to $4 trillion by 2030. Speaking at the Goldman Sachs Communacopia + Technology Conference, Huang said demand is growing faster than supply, with NVIDIA targeting 70% annual revenue growth while describing unconstrained demand growth as exceeding 100%. He cited supply-chain constraints in packaging, DRAM, connectors, voltage regulators and wafers, as well as downstream limits in land, power and data-center buildings, and said the company has a year to expand supply. Huang named CoreWeave, Nebius, Nscale, Lambda and Firmus as regional cloud providers helping secure capacity beyond the largest cloud service providers, and said NVIDIA announced work in Australia involving 2 gigawatts of capacity for 2027, representing approximately $80 billion in infrastructure. He also pointed to cybersecurity, drug discovery and physical AI as major growth opportunities, citing partnerships with CrowdStrike, Cisco and Palantir, and said NVIDIA has $100 billion in lined-up contracts.

NVIDIA has handed the software running the 1.3 million-part supply chain inside every Vera Rubin rack to Palantir, expanding a partnership that fuses Palantir Foundry and AIP with NVIDIA Nemotron open models grounded in Palantir Ontology. The reference architecture, called SAIOS, runs on hardware from Dell Technologies and Cisco for on-premises deployment, while Rackspace and Nebius enable co-location and cloud options. NVIDIA CEO Jensen Huang called supply chains "the operating system of the physical economy," and Palantir CEO Alex Karp said the sovereign stack is "delivering capabilities that exceed the frontier while providing alpha protection qualities unavailable otherwise." The endorsement lands as Palantir shares traded at $167.15 midday Thursday, down 1.41% on the session, and are down 6.2% year to date even as NVIDIA has climbed 17.52%. Palantir's Q2 FY2026 revenue reached a record $1.94 billion, up 92.83% year over year, with adjusted EPS of $0.41 versus a $0.28 consensus, U.S. commercial revenue up 149% to $764 million, and full-year revenue guidance raised to $8.150 to $8.158 billion, yet the stock carries a P/E of 240 and a market cap near $383.9 billion.

Palantir Technologies and NVIDIA announced a collaboration to bring sovereign AI to critical supply chains, starting with NVIDIA's own operations. The deployment creates an AI stack that brings NVIDIA Nemotron open models into Palantir Foundry and Artificial Intelligence Platform, grounded in the Palantir Ontology, aimed at creating supply-chain visibility, identifying constraints and guiding decisions at machine speed. NVIDIA operates one of the most complex supply chains, spanning millions of parts, thousands of suppliers and a global network of manufacturing partners, with 1.3 million parts in each NVIDIA Vera Rubin rack. The deployment runs on NVIDIA reference architectures and the jointly developed Palantir Sovereign AI Operating System Reference Architecture, supported by Dell Technologies and Cisco, and can be deployed on premises with Cisco and Dell or in the co-lo and cloud with Rackspace and Nebius. The AI stack and its applications across industries will be showcased at Palantir's AIPCon 11.

Nebius Group N.V.'s shares jumped 7.73% to close at $243.88 after Palantir Technologies partnered with Nebius for AI-native cloud infrastructure, designating Nebius as its preferred sovereign AI infrastructure partner. Under the agreement, Nebius will provide AI compute and inference infrastructure to Palantir's commercial customers, with endpoints expected to operate inside Palantir's enterprise perimeter. The partnership aims to offer an end-to-end platform for running open AI models and training on proprietary data, with plans to accelerate capacity deployment through modular data centers. Nebius reported $582 million in second-quarter revenue, up 454% year over year, and maintained its 2026 revenue outlook of $3-$3.4 billion, expecting connected power capacity to reach roughly 800 MW to 1 GW by end of 2026. The company faces tough competition from CoreWeave and Microsoft, but Palantir's endorsement could provide a key distribution channel and reduce customer-acquisition friction.

Intel and most of the semiconductor sector rose on Tuesday, while the broader market dipped, with the Dow down 1%, the S&P 500 down 0.3%, and the Nasdaq Composite down 0.1%. The Philadelphia Semiconductor Index gained 2.2%. Intel shares jumped 9% after Northland Securities upgraded the stock to Outperform and set a $120 price target, citing a reported plan to raise CPU prices by up to 10% and a partnership with Tesla on the Terafab semiconductor endeavor. Qualcomm rose 4% after announcing a partnership with Amazon to build next-generation AI data center infrastructure, issuing warrants for up to 25 million shares at $161.26 per share. AMD climbed 6%, Taiwan Semiconductor Manufacturing gained 2.5%, and GlobalFoundries rose 2.3% after securing a $375 million award from the U.S. Department of Commerce's CHIPS Research and Development Office. Broadcom gained 3.7%, Marvell rose 2.8%, Arm Holdings gained 3.5%, and MaxLinear surged 7%. Among neoclouds, Nebius soared 10% after Palantir named it preferred AI infrastructure partner, CoreWeave surged 14%, and IREN climbed nearly 10%. Microsoft declined 1.4%, Amazon inched down 0.6%, and Oracle increased 3.5%.

Antioch, a simulation platform for physical AI, announced a $32 million Series A funding round led by Greylock, with participation from A*, Category Ventures, Box Group, and Icehouse Ventures. The company, founded by alumni of Tesla and Meta Reality Labs, enables robotics and autonomy teams to test AI systems in high-fidelity simulations before real-world deployment. Antioch is already working with Amazon's Ring and has partnerships with NVIDIA and Nebius. The funding will accelerate product development and expand engineering efforts as demand grows across robotics, drones, and industrial automation.

Bloom Energy has officially earned a spot in the S&P 500, with S&P Dow Jones announcing on September 4 that the company, along with Illumina and Everpure, will join the index, while Molson Coors, Builders FirstSource, and The Trade Desk were removed. The stock jumped after hours on the news and is set to join the index on September 21. Bloom Energy has been a primary beneficiary of the huge power demand from AI data centers, addressing the bottleneck of grid connection delays with its solid-oxide fuel cells that can be installed on-site in three to four months. In the second quarter, revenue exceeded $1 billion for the first time, and management raised its full-year revenue projection to a range of $3.9 billion to $4.2 billion, a 100% increase year over year at the midpoint. Brookfield Asset Management expanded its partnership, increasing its project financing framework from $5 billion to $25 billion, and a separate $1.7 billion project investment was announced to support Nebius' AI infrastructure. Analysts project earnings per share to rise to $2.67, a 251% increase from 2025's EPS of $0.76, though the stock trades at about 94 times projected earnings for this year.

Nvidia CEO Jensen Huang said at the G20 Summit that building a 1-gigawatt AI facility costs between $50 billion and $60 billion, a figure that underscores the scale of the AI infrastructure buildout and has implications for neocloud providers like Nebius. Nebius, which reported Q2 revenue of $582 million, up 454% year over year, and annual recurring revenue of $3 billion, aims for $7 billion to $9 billion in annual recurring revenue by year-end. Huang's valuation suggests Nebius' projected five gigawatts of contracted power by end-2026 could be worth up to $300 billion once fully built, though that does not imply an immediate market cap of that size. The higher valuation also strengthens neoclouds' ability to secure customer prepayments, which already fund 50% to 60% of Nebius' capital expenditures, and could enable future borrowing against data centers, as Iren CEO Dan Roberts noted that his company's portfolio remains unencumbered. These factors address the bearish concern about how neoclouds will finance their massive buildouts.

NVIDIA Corporation's Groq 3 LPX rack is in full production and will be deployed alongside its Vera CPUs and Rubin GPUs at neocloud Nebius later this year, according to Nvidia senior director Dion Harris. The rack marks the commercialization of technology from Nvidia's $20 billion purchase of Groq's assets in December, its largest deal on record. Each liquid-cooled rack packages 256 Groq chips and can deliver 3,400 tokens per second, a benchmark from Artificial Analysis. Groq's chips are manufactured by Samsung, unlike Nvidia's own GPUs, which are made by Taiwan Semiconductor Manufacturing Co. The rollout currently relies on Nebius as the only confirmed customer, and Nvidia has also committed up to $100 billion to OpenAI and $5 billion to Intel, among other investments.

Nebius Group N.V. is positioning itself for strong growth in 2027, driven by large AI infrastructure deals and expanding capacity. In the second quarter of 2026, the company signed four landmark AI cloud agreements with Reflection, Cohere, a scaled U.S. new lab, and a large U.S.-based quantitative trading firm, each averaging over $1 billion and yielding roughly $20-$25 million per megawatt. Customer prepayments cover 50-60% of associated capital expenditures. Management expects these contracts to contribute meaningfully to revenues in 2027, though not materially in 2026. Nebius raised its year-end 2026 contracted power target to 5 gigawatts and plans to add over 1 gigawatt of new capacity annually, with deployment targeted for 2027. The company has deliberately not pre-sold all planned 2027 capacity, aiming to capture better pricing and serve short-term, higher-margin demand, and it could already have sold out that capacity. Nebius also sees opportunities in shorter-duration agreements with premiums of roughly $40-$50 million per megawatt. With strong pricing, sizable commitments, and greater capacity, Nebius appears well-positioned for rapid expansion in 2027.