NavigationContentFooter

GPU Instances Documentation

GPU Instances

Scaleway GPU Instances are equipped with dedicated high-end NVIDIA GPUs and offer the speed and efficiency you need to run your most demanding workloads.

Instances Quickstart

Getting Started

Quickstart

Learn how to create, connect to, and delete a GPU Instance in a few steps.

View Quickstart

Concepts

Core concepts that give you a better understanding of Scaleway GPU Instances.

View concepts

How-tos

Check our guides to creating and managing GPU Instances and their features.

View how-tos

Additional content

Guides to help you choose a GPU Instance, understand pricing and advanced configuration.

View additional content

Tutorials

Deploying models with NVIDIA Triton Inference Server on Scaleway Object StorageLink
Achieving Super-Resolution with a Sub-Pixel Convolutional Neural Network on Scaleway GPULink
Instance API

Learn how to create and manage your Scaleway GPU Instances through the API.

Go to Instances API

Changelog

  • GPU Instances

    Added

    New GPU L4 Instances

    We are excited to introduce our latest versatile and cost-effective GPU Instances based on Nvidia L4 in the Paris (PAR2) and Warsaw (WAW2) regions.

    Key features include:

    • Nvidia L4 24 GB (Ada Lovelace architecture)
    • 4th generation Tensor cores
    • 4th generation RT cores (graphics capability)
    • Available in multi-GPU 1, 2, 4 or 8 GPUs
  • GPU Instances

    Changed

    H100 PCIe GPU Instances now available in Poland

    We are excited to expand the availability of our H100 Instances, to provide you with the most high-end AI chip of the market! Scaleway's H100 Instances based on Nvidia H100 Tensor Core GPUs, are now available in the Warsaw (WAW2) region.

    Key features include:

    • Nvidia H100 80GB (Hopper architecture)
    • 4th generation Tensor cores
    • Multi-Instance GPU (MIG)
    • Transformer Engine
    • Available in multi-GPU 1 or 2 GPUs
  • GPU Instances

    Added

    New H100 PCIe GPU Instance

    We are excited to introduce our latest GPU Instances based on Nvidia H100 Tensor Core GPU, the most high-end chip on the market, in the Paris (PAR2) region.

    Key features include:

    • Nvidia H100 80GB (Hopper architecture)
    • 4th generation Tensor cores
    • Multi-Instance GPU (MIG)
    • Transformer Engine
    • Available in multi-gpu 1 or 2 GPUs
View the full changelog
Questions?

Visit our Help Center and find the answers to your most frequent questions.

Visit Help Center
Docs APIScaleway consoleDedibox consoleScaleway LearningScaleway.comPricingBlogCarreer
© 2023-2024 – Scaleway