Neocloud

Nebius

The ex-Yandex team's AI cloud: aggressive pricing on big self-built European and US fleets.

NASDAQ-listed, building its own datacenters and servers, with H100/H200/B200 fleets and Anthropic-scale customers... plus a managed inference and fine-tuning layer (Nebius AI Studio). Frequently the price leader among 'serious' neoclouds.

Visit Nebius ↗

GPU pricing on Nebius

On-demand list per GPU-hour, last reviewed 2026-09. Methodology.

GPU $/GPU-hr vs cheapest Notes
A100 80GB $1.15 cheapest
H100 SXM $1.95 1.7×
H200 $2.30 2.0×
B200 $4.80 4.2×

Strengths

  • Among the lowest H100/H200 on-demand prices
  • Self-designed hardware, real capacity
  • EU data residency options

Watch for

  • Younger US footprint
  • Ecosystem still maturing

Alternatives

CoreWeave

The flagship neocloud: Kubernetes-native GPU fleets built for training clusters, now public company.

Lambda

The ML researcher's default: one-click clusters, sane pricing, no quota theater.

RunPod

Community and secure cloud tiers, per-second billing, and serverless GPU endpoints.