From Open Source to Real-World Scale
summary
Nebius describes a vertically integrated AI cloud that promises click-to-provision access to thousands of GPUs with a claim of fully virtualized physical GPU performance, backed by a NASDAQ listing and Nvidia reference/exemplar status. Core stack pieces named: KVM, Kubernetes, VPP networking, YDB storage, and an open-sourced Slurm operator called s-operator. Engineering focus emphasizes speed (instant clusters), stability under real workloads, pay-as-you-go plus reserved capacity, and hands-on support; demos and credits are available to validate the claims.
AI-written summary · about
related talks
17mAI Foundations: The Power of Infrastructure
22mScaling agentic AI: From infrastructure to enterprise outcomes – Duncan Ng
01h 15mTutorial: Build, Operate, and Use a Multi-Tenant AI Cluster Based Entirely on Open Source
21mAnalyst Notebook
04mA Cloud Native Stack From Bare Metal To Tokens for Large-Scale AI Inference
16mHyperbolic
05mSponsored Keynote: Towards Clouds of AI Clusters
28mScaling agentic AI: From infrastructure to enterprise outcomes