Certifications such as SOC 2 Type II including HIPAA, and ISO 27001 enable AI deployment in highly regulated contexts including healthcare, financial services, and government
Enhanced observability and Identity and Access Management (IAM) provide more granular control, while streamlined developer workflows accelerate time to production with simplified UI and app catalogue
Platform resilience further strengthened with proactive system health monitoring, self-healing infrastructure, and substantial performance gains across storage systems.

Nebius today announced Nebius AI Cloud 3.0 "Aether", the latest release of Nebius AI Cloud, delivering enterprise-grade features that give organizations the trust, control, and simplicity they need to run their most critical AI workloads in production and at scale.

As AI workloads shift from training to inference at scale, organizations need a complete development platform that accelerates time-to-value. Yet existing cloud infrastructure often falls short: lengthy procurement delays deployment, fragmented tooling slows development, and inconsistent performance across regions creates complexity. Security models built for general-purpose workloads can't meet AI-specific compliance requirements, while managing data pipelines, access controls, and GPU resources creates bottlenecks.

With the Aether release, Nebius AI Cloud delivers what enterprises need for production AI: independently validated security certifications (SOC 2 Type II including HIPAA, ISO 27001, ISO 27799) plus alignment with NIS2, DORA, ISO 27032, and ISO 27701 principles; fine-grained governance with native observability, granular IAM, and full infrastructure visibility; and streamlined developer workflows that accelerate deployment. These capabilities enable healthcare, financial services, government, and other regulated organizations to move AI workloads into production with confidence.

Performance and reliability remain foundational. Nebius recorded leading results in the MLPerf® Inference v5.1 benchmarks with NVIDIA GB200 NVL72 and NVIDIA HGX™ B200 systems and became one of the first NVIDIA Exemplar Clouds, achieving NVIDIA performance standards for large-scale training workloads on NVIDIA HGX H200. The Aether release adds automated health monitoring with proactive issue detection and significant storage performance upgrades to ensure enterprise-scale stability.

"Nebius AI Cloud delivers what enterprises need: supercomputer performance, hyperscaler ease-of-use, and the trust and governance required for production AI," said Roman Chernin, co-founder and Chief Business Officer of Nebius. "Aether proves that enterprise control and builder productivity can coexist on the same platform. With comprehensive security and governance features, and a constant focus on the developer experience, we're delivering the infrastructure enterprises need to move from AI experimentation to business-critical production."

Nebius is building out a comprehensive ecosystem to accelerate enterprise AI adoption. Partnerships with technology distributors such as TD SYNNEX, which launched its AI Infrastructure as a Service offering in North America on Nebius AI Cloud, and integrations with platforms including Anyscale, SkyPilot, Outerbounds, Saturn Cloud, and Lightning AI provide end-to-end solutions from procurement to deployment.

With infrastructure deployed across the US, Europe, the UK, and Middle East, Nebius AI Cloud offers global compliance and local data residency options that meet regional requirements. Combined with early access to the latest GPU architectures, enterprises can deploy AI workloads locally with a unified platform experience.

Nebius will hold a walk-through of Nebius AI Cloud 3.0 "Aether" including a live Q&A session on Wednesday, 5 November, at 6 p.m. CET / midday EST / 9 a.m. PST. Please register here for more information.

Nebius AI Cloud 3.0 "Aether" – at a glance

Enterprise-grade security & compliance

SOC 2 Type II, including HIPAA section, and ISO 27001 certifications validated by independent auditors
Alignment with NIS2, DORA, ISO 27032, ISO 27701 and ISO 27799 regulatory frameworks
Embedded secrets manager for secure API key and credential management

Governance & control

Enhanced IAM feature set with more granular controls
Comprehensive observability suite with searchable logs, histograms, metrics ingestion, and auditing

Developer productivity

Updated UI and navigation that better align with AI ML developer workflows
Support for VM creation with preloaded apps for rapid AI workload deployment
Simplified GPU, CPU and memory quota allocation and billing
Expanded ecosystem support including SkyPilot Server and Anyscale integrations

Reliability & performance

Automated active and passive health checks with automatic node repair
Object and file storage performance enhancements (third-party and native options)
NVIDIA Exemplar Cloud on NVIDIA HGX B200
Leading results in MLCommons MLPerf® Inference v5.1 benchmarks with NVIDIA GB200 NVL72 and NVIDIA HGX NVIDIA B200 systems

Additional resources

Blog post from our Product Management team
Blog post on enterprise security certifications
Webinar and live Q&A registration
Nebius Trust Center

About Nebius

Nebius is a technology company building full-stack cloud infrastructure for the global AI industry. Headquartered in Amsterdam and listed on Nasdaq (NASDAQ:NBIS), the company has a global footprint with R&D hubs across Europe, North America, and Israel.

Nebius AI Cloud has been built from the ground up for intensive AI workloads. With proprietary software and hardware designed in-house, Nebius AI Cloud gives AI builders the compute, storage, managed services, and tools they need to build, tune, and run their models.

View source version on businesswire.com: https://www.businesswire.com/news/home/20251016822814/en/

Investor Relations:

askIR@nebius.com

Media Relations:

media@nebius.com