Nebius · Pricing Plans
Nebius Plans Pricing
Nebius Plans Pricing is the pricing profile for Nebius on the APIs.io network — every figure transcribed verbatim from the provider’s own pricing page on 2026-10-07.
It defines 30 plans, including NVIDIA GB300 NVL72 — Contact us; NVIDIA HGX B300 — $7.85; NVIDIA GB200 NVL72 — Contact us, and 27 more.
The page prices API or platform usage directly, priced by usage; the service is delivered as managed.
Tagged areas include Artificial Intelligence, Cloud, Compute, GPU, and HPC.
30 Plans
Prices the API
Basis: usage other
Delivered: managed
Transcribed from the provider's pricing page on 2026-10-07 — nebius.com/prices.
View Source
Artificial IntelligenceCloudComputeGPUHPCInferenceKubernetesMachine LearningStorage
Plans
NVIDIA GB300 NVL72
none
Contact us per GPU-hour
Includes: 112 vCPUs, 800 GB RAM
NVIDIA HGX B300
per-use
$7.85 per GPU-hour (on-demand)
Includes: 24 vCPUs, 346 GB RAM
Overage: $9.50 GPU-hour effective October 1, 2026
NVIDIA GB200 NVL72
none
Contact us per GPU-hour
Includes: 112 vCPUs, 800 GB RAM
NVIDIA HGX B200
per-use
$7.15 per GPU-hour (on-demand)
Includes: 20 vCPUs, 224 GB RAM
Overage: $8.50 effective October 1, 2026
NVIDIA HGX H200
per-use
$4.50 per GPU-hour (on-demand)
Includes: 16 vCPUs, 200 GB RAM
Overage: $5.40 effective October 1, 2026
NVIDIA HGX H100
per-use
$3.85 per GPU-hour (on-demand)
Includes: 16 vCPUs, 200 GB RAM
Overage: $4.50 effective October 1, 2026
NVIDIA RTX PRO 6000
per-use
$1.80 per GPU-hour
Includes: 24 vCPUs, 218 GB RAM
NVIDIA L40S with Intel CPU
per-use
from $1.55 per GPU-hour
Includes: 8-40 vCPUs, 32-160 GB RAM
NVIDIA L40S with AMD CPU
per-use
from $1.82 per GPU-hour
Includes: 16-192 vCPUs, 96-1152 GB RAM
Preemptible NVIDIA HGX B300 / B200
per-use
from $0.99 per GPU-hour spot
Preemptible NVIDIA HGX H200 / H100 / RTX PRO 6000
per-use
from $0.79 per GPU-hour spot
Preemptible NVIDIA L40S with Intel CPU
per-use
from $0.74** per GPU-hour
Preemptible NVIDIA L40S with AMD CPU
per-use
from $0.90** per GPU-hour
AMD EPYC Genoa (CPU-only)
per-use
from $0.10 per hour
Includes: 4-64 vCPUs, 16-256 GB RAM
Overage: from $0.13 effective October 1, 2026
Intel Ice Lake (CPU-only)
per-use
from $0.05 per hour
Includes: 2-80 vCPUs, 8-320 GB RAM
Overage: from $0.06 effective October 1, 2026
Shared Filesystem (by Nebius)
month
$0.0800 per GiB/month
WEKA filesystem
month
$0.1000 per GiB/month
Object storage — Standard (volume)
month
$0.0147 per GiB/month
Object storage — Standard (egress traffic)
per-use
$0.0150 per GiB
Object storage — Intelligent, Warm Tier (volume)
month
$0.030 per GiB/month
Object storage — Intelligent, Cold Tier (volume)
month
$0.010 per GiB/month
Object storage — Enhanced (volume)
month
$0.1100 per GiB/month
Block volumes without data replication
month
$0.0530 per GiB/month
Block volumes with erasure coding
month
$0.0710 per GiB/month
Block volumes with 3x mirroring
month
$0.1180 per GiB/month
Block volumes — local SSD disk
month
$0.065 per GiB/month
Managed Service for Kubernetes
none
Free
Managed Soperator (Slurm on Kubernetes)
none
Free
Includes: Free software, consumption-based pricing
Networking: Egress/Ingress traffic
none
Free
Networking: Public IP address
none
Free
Sources
Work with this as data
Every plan here is available over the APIs.io API and to AI agents over MCP.