Gcore

What's good about Gcore

• Wide global edge presence beyond typical neocloud regions
• Access to Blackwell Ultra B300 and GB300 NVL72 systems
• Combined CDN, networking and GPU stack from a single provider
• Virtual GPU shelving preserves IPs and volumes at zero compute cost between runs

Gcore GPU prices

5 GPU models across 20 configurations. Rank compares Gcore's cheapest on-demand listing for each GPU with every provider's cheapest.

What a run costs on Gcore

24 hours at today's cheapest listing for each GPU, at the size it is listed.

• 1× H100 SXM for 24 hours

  Everywhere Inference at $1.72/GPU-hr

  $41.18

  $35.62 under the market median rate

• 8× H200 for 24 hours

  Bare Metal at $3.57/GPU-hr

  $684.96

  $186.68 under the market median rate

• 1× A100 SXM for 24 hours

  Everywhere Inference at $1.41/GPU-hr

  $33.91

  $4.49 under the market median rate

The market median is every provider's cheapest on-demand price for that GPU.

Billing & terms

Virtual GPU

Hourly billing for self-serve GPU virtual machines, charged only while the instance is powered on

Bare-Metal GPU Clusters

Dedicated 8-GPU servers and multi-node clusters, priced per configuration with hourly, monthly and yearly estimates in the GPU Cloud calculator

Everywhere Inference

Billing determined by the exact compute time used, measured down to the minute

Quote-Based Capacity

B300 and GB300 NVL72 systems, plus third-party cloud and on-premises deployments, are quoted by the sales team

Regions

50+ cloud locations and 210+ points of presence worldwide; bare-metal GPU clusters are offered in a smaller subset of regions such as Luxembourg

Support

Documentation, API reference, help center, status page and technical support plans

Watch out for

• Catalog spans many products beyond GPU compute, which can complicate pricing
• GPU SKU availability varies meaningfully by region
• Less GPU-specialized than dedicated training neoclouds
• Newest Blackwell Ultra SKUs are quote-based rather than self-serve
• Published prices are denominated in EUR and exclude VAT

Features

Global Edge Footprint

GPU and inference workloads deployable across 50+ cloud locations and 210+ points of presence worldwide

Blackwell Ultra Systems

8-GPU B300 SXM servers and 72-GPU GB300 NVL72 racks offered alongside H200, H100, A100 and L40S

Everywhere Inference

Managed inference platform with autoscaling, fast cold starts and protected endpoints, billed on the compute time actually used

Bare-Metal and Virtual GPUs

Dedicated 8-GPU bare-metal clusters or single-GPU virtual machines, with multi-node clusters provisioned in a single API call and automatic InfiniBand setup

Private and Third-Party Deployments

AI Cloud Stack and Everywhere AI for running models on enterprise, third-party cloud or on-premises infrastructure

Services

• GPU Cloud

  Bare-metal GPU clusters and virtual GPU machines provisioned through the console or API

• Everywhere Inference

  Managed endpoints for running models across Gcore's global footprint

Getting started

1. Step 1

   Create an account

   Sign up at the Gcore customer portal

2. Step 2

   Select GPU product

   Choose between cloud GPU instances, bare-metal clusters, or the inference platform

3. Step 3

   Provision and connect

   Launch via the portal or API and connect over SSH or supported orchestrators