Nebius
AI overview
Sign in and the AI will write an overview from our coverage.
Headlines · 47
- Nscale Joins PyTorch Foundation as Platinum Member
Nscale has joined the PyTorch Foundation as a Platinum member, gaining a seat on the organization’s governing board. The AI cloud provider said it will continue investing in Ray following its planned acquisition of Anyscale and contribute to open-source infrastructure for distributed AI workloads.
PyTorch Blog · 🔥 0 - NVIDIA Vera CPUs draw Nebius, Nscale and other European infrastructure partners
NVIDIA's blog said Nebius, Nscale and Verda are among global technology partners building the next wave of infrastructure around its Vera CPUs, aimed at agentic computing workloads. NVIDIA says agents retrieve information, run software tools, execute code and evaluate results, and claims each step depends on fast CPU processing, with Vera designed to sustain that speed while running thousands of tasks simultaneously. Server makers Dell Technologies, HPE, Lenovo, Supermicro and Bull are building Vera-based systems, and NVIDIA says data-processing companies including SAP, ClickHouse, Cloudera, Elastic and Starburst observed strong results on Vera in early testing for analytical, extract-transform-and-load and semantic search workloads.
NVIDIA Blog · 🔥 0 - EBU and NVIDIA Expand Sovereign AI Framework for European Public-Service Media
The European Broadcasting Union and NVIDIA are expanding a sovereign framework for deploying AI agents in European public-service media, adding cloud providers, open-model developers and systems integrators. The initiative is intended to define requirements and reference architectures for workload placement, data governance, model selection and lifecycle security while broadcasters retain control of their data, infrastructure, models and security. Microsoft, Nebius, Qvest and NVIDIA Nemotron open models are among the components described in the report. Adobe also presented Frame.io enhancements using NVIDIA GPUs for semantic video search, reporting about 5x faster embedding generation, up to 4x higher throughput and up to 5x lower search latency. The report does not specify a rollout timetable or confirm that every named participant will be included in the initial deployment.
NVIDIA Blog · 🔥 0 - Neocloud shakeout looms, with Nvidia exposed to AI cloud consolidation
The Information reports that after Firmus abruptly pulled its IPO and shares of CoreWeave and smaller rivals fell about 20%, data center and neocloud executives expect only seven to ten of hundreds of AI cloud providers to still be around in a few years, with consolidation seen as inevitable. The Decoder adds that customers outside the hyperscalers accounted for 45% of Nvidia's data center revenue—$40 billion in the July quarter—and that Nvidia is backing partners with equity investments, chip allocations, loan guarantees and help finding sites; Gartner predicts 80% of independent neoclouds will cease independent operations by 2030. The claim that Microsoft pulled out of a planned Nscale data center lease in England comes from The Telegraph, and Nscale says its Microsoft relationship remains strong and long-term.
The Information · 🔥 0 - Microsoft pulls out of Nscale's UK data center as grid power slips to the 2030s
Nscale said Microsoft has withdrawn from the Loughton, Essex data center the British neocloud firm was building for it, after UK grid authorities told Nscale not to expect a power supply until the 2030s; the site had been designed to host 23,040 Nvidia GB300 GPUs, starting at 50 MW and expandable to 90 MW. Microsoft was the only publicly announced customer, and Nscale says it remains committed to the project, is exploring alternative power sources and is looking at other UK expansion sites while calling its Microsoft relationship strong and long-term. The exit comes as Nscale seeks to raise as much as $3 billion in a US IPO, with Microsoft and Anthropic accounting for about 85% of its contracts, and Bloomberg reported investors pushing for a valuation of $16–19 billion against the roughly $30 billion Nscale had sought.
The Information · 🔥 0 - DeepSeek releases V4.1-Flash-1017 open-weight model
DeepSeek released DeepSeek-V4.1-Flash-1017, which the company described as a Pareto improvement in performance and cost, now live on its website, app and API, with a testing account on X saying it is also on HuggingFace. 36Kr reports the base architecture is unchanged and the update focuses on capability iteration: DeepSWE v1.1 Pass@1 rose from 74.2 to 77.0, above Claude Opus 5 at 74.0 and GPT-5.6 Sol at 73.0, and Terminal-Bench 4.0 went from 31.2 to 42.3, while HLE scored 42.6 versus 56.3 for Opus 5. The report adds that API access and pricing are unchanged, with savings coming from shorter task runtimes and outputs, and that a Nebius employee's independent test passed all 30 runs versus 24 of 30 for the earlier version, at about 4.5 cents and roughly 550,000 tokens per task.
testingcatalog (X) · 🔥 0 - Gartner: $3.5tn of AI infrastructure spending at write-down risk by 2030
Gartner's analysis published on 14 October estimates that more than $3.5 trillion of AI infrastructure spending could face impairment by the end of 2030 if providers keep investing at the current pace, with more than $4 trillion of asset value at write-down risk through 2032. The firm expects nearly $1 trillion of AI infrastructure spending in 2026 and about $6 trillion more between 2027 and 2030, while AI data centre owners would need $8.5 trillion in revenue to cover that spending, financing costs and a 12% return on invested capital, against forecast business and consumer AI revenue of no more than $3.6 trillion. Gartner also predicts that more than 80% of neoclouds — companies that rent out AI computing power — will stop operating independently by 2030, citing cost pressure and a lack of business-model differentiation, while hyperscalers with diversified revenue are better placed to sustain AI investment.
Bloomberg Technology · 🔥 0 - CoreWeave shares and bonds slide as investors sour on AI infrastructure debt
CoreWeave closed down about 15.5% on Oct 14 as yields on its bonds approached 15% amid investor concern about heavy AI-related borrowing, with Oracle down 8.8%, Nebius 10% and Dell 11.3% in the same session. Reuters reported the stock was down more than 15% for the week by Thursday's close, the yield on one bond due in 2031 briefly rose above 16%, and the cost of insuring its debt against default climbed past 1,000 basis points, a record. Bloomberg, cited by 36Kr, reported the three co-founders have sold $3.3 billion of stock since the March 2025 IPO, while TechNews said CEO Michael Intrator and related entities had sold over 10 million shares for nearly $1 billion; the company says its signed backlog of more than $129 billion covers its debt, and in August it raised its full-year capex forecast to $35–39 billion.
Techmeme · 🔥 0 - theCUBE Pod: AI doubters get louder as IBM bets on commodity models
On the latest theCUBE Pod, SiliconANGLE's John Furrier and Dave Vellante discussed public pushback against AI, IBM's AI strategy and the industry's shift from training toward inference. The episode noted that shares of Nvidia, CoreWeave, Oracle and Nebius fell last week amid reports that OpenAI's annualized net revenue was about $50 billion, roughly $20 billion below previously reported figures. Furrier called the uproar an "AI freakout moment" and said it is temporary, while Vellante argued customers first spend on infrastructure and models and then turn to IBM to put that intelligence to work, which poses a challenge for the company.
SiliconANGLE AI · 🔥 0 - NVIDIA open-sources full Run:ai platform under DSX AI Factory
NVIDIA said on Oct. 13 that it is releasing the complete NVIDIA Run:ai platform as open source software as part of its DSX AI Factory platform, including the self-hosted control plane and four components that can also be adopted separately: KAI Scheduler, GPU Fractioning Operator, KAI Resource Management and Workload Map. TechRepublic reported the codebase is published under the Apache 2.0 license, letting teams deploy it in production without a Run subscription, while enterprise support stays paid and free users handle their own infrastructure, upgrades and operational issues. DoNews said eight DSX ecosystem partners, including Crusoe, Mirantis, Nebius and Rafay, have integrated the open-source release, and cited a loading benchmark on eight H100 VMs that other reports did not corroborate.
NVIDIA Technical Blog · 🔥 0 - Amazon reportedly in talks to buy Decart for about $7B
The Wall Street Journal reported that Amazon is negotiating to acquire San Francisco AI startup Decart in a deal that could be valued at around $7 billion, a report relayed by Bloomberg, IT之家 and others. Founded in 2023, Decart sells a cross-chip optimization layer called DOS that lets AI workloads move across Nvidia GPUs, Amazon Trainium and Google TPU, and also builds the real-time video model Lucy and the interactive-environment model Oasis. Reports say Anthropic discussed a roughly $6 billion acquisition in August and walked away after due diligence, while Nvidia was also named as a possible suitor; Amazon and Decart did not comment, and TechRepublic noted CTech floated a possible $5.5 billion valuation.
The Next Web · 🔥 0 - OpenAI's annualized revenue said to be closer to $50B, not $70B
The Information reports that the widely circulated figure of $70 billion in OpenAI annualized revenue was far off, and that the company is generating closer to $50 billion. After a Financial Times report of the lower number on Thursday, shares of chip firms and neoclouds including Nvidia, CoreWeave and Nebius fell. The piece argues that multiplying one month's revenue by 12 is a poor substitute for a fuller picture of a company's performance.
The Information · 🔥 0 - Sesterce plans €10B AI data centre campus at former Finnish paper mill
French AI infrastructure company Sesterce said it plans to invest more than €10bn in an AI data centre campus in Jämsä, central Finland, on the Kaipola site where UPM shut its paper mill in early 2021. The first phase, at 200MW, is due to start construction in 2026, with a second phase lifting capacity to 600MW; over time Sesterce aims for more than 1GW of AI capacity on existing industrial sites in Finland. The company expects around 2,000 construction jobs and about 300 permanent roles, plus a €10m local fund; site owner Kaipola Green Port is still negotiating the sale, and local officials note permits are still required.
36氪快讯 · 🔥 0 - Nebius, NVIDIA and Uber on squeezing more value from every unit of compute
At a panel hosted by The Information's AI Agenda Live, executives from Nebius, NVIDIA and Uber discussed how they are optimizing compute as agentic workflows threaten to multiply token consumption. Nebius CRO Marc Boroditsky said the company tailors optimizations in its Token Factory inference platform to each customer's goals—speed, reliability, quality or price—using techniques such as speculative decoding, where smaller models predict likely next tokens that a larger model verifies in bulk, tuned with the customer's own traffic. NVIDIA's Dion Harris pointed to a Dynamo feature that locates already-computed data in a cluster's KV cache so teams avoid recalculating it, while Uber infrastructure VP Mattie Toia said its token costs have stabilized despite growing agentic use, helped by caching pushed down to individual sub-agents and by giving engineers visibility into their own usage; Uber also built an internal "context graph" of roughly 24 million nodes that cut query times from 20 minutes to 30 seconds.
The Information · 🔥 0 - AI neocloud Lambda to raise $4B at $14.5B valuation ahead of 2027 IPO
Cloud provider Lambda is raising up to $4 billion at a $14.5 billion pre-money valuation, led by Coatue Management and Blackstone, in what could be its last private round before a planned 2027 IPO, according to The Wall Street Journal. TechCrunch, citing an investor letter, reports Lambda's backlog grew from $15 billion in June to $50 billion in September, with much of that increase driven by a $35 billion commitment from Anthropic signed in late August. Lambda also closed an additional $1 billion in debt financing last week to buy more chips.
TechCrunch AI · 🔥 0 - Reflection AI launches Beam, a 501B open-weight model
Reflection AI unveiled Beam on October 5, its first open-weight model, aimed at coding, reasoning and agentic workloads. Beam is a sparse mixture-of-experts model with 501 billion total parameters and 23 billion active, pretrained on 23.8 trillion tokens with a 1 million-token context window. The company says it approaches GLM-5.2 on some advanced reasoning tasks using roughly one-third to one-quarter of the inference compute, but the claims have not been independently verified; weights, a technical report and tools are slated for release later in October.
IT之家 AI · 🔥 0 - Nebius’ Eigen AI acquisition puts inference efficiency at center
Nebius acquired inference-optimization startup Eigen AI for consideration exceeding $1 billion, bringing its roughly 20-person team into the cloud provider. Eigen AI founder Hanrui Wang now leads Nebius’s Token Factory, which covers inference, post-training and agent systems, while the company positions efficient open-model serving as a core business.
MIT科技评论中文 · 🔥 0
Experience and discussion from the community
Share my Nebius experienceAsk about Nebius
Nobody has shared their experience with Nebius yet.