Buy Scaleway Account

Is a Scaleway account the shortest route to a fully European cloud stack?

For a growing number of teams the answer is yes, and the fastest way to test that answer is to buy Scaleway account online and spend a week inside the console. Scaleway is the cloud arm of the Iliad Group, headquartered in Paris, with data centers in France, the Netherlands and Poland and a catalog that reaches from bare metal to managed Kubernetes, serverless functions and NVIDIA GPU instances. AllCloudShop supplies a single variant on this listing: a Free Trial with $100 Credit, priced at $45, which is exactly what its name says and nothing more.

You are probably here for one of three reasons. You may want compute that lives entirely inside the European Union, under a provider that is itself European. You may want a managed Kubernetes product that does not require a hyperscaler bill to run. Or you may want GPU capacity for model training and inference without the waiting lists that bigger clouds impose. Scaleway answers all three, and a Scaleway account for sale on this page is the entry point to every one of those services.

What follows is a working guide rather than a sales pitch. It explains the platform's Organization and Project model, the IAM system, the scw command-line tool, the three regions and their availability zones, the Instance families, Elastic Metal, Kapsule, Object Storage, the Serverless products, how the credit balance is applied, the traps that catch new users, and where Scaleway sits beside OVHcloud, Hetzner, DigitalOcean and Civo.

Buyers who want to buy verified Scaleway account access should read the variant section closely, since the phrase "verified" is used here only to mean an account that has completed Scaleway's ordinary sign-up process and is in an active state.

If anything about the listing is unclear after reading, or you simply want to buy Scaleway account online without a long pre-sales exchange, the Telegram and WhatsApp channels on the contact page exist for exactly that purpose. Ask before you order rather than after.

Scaleway's catalog, from Paris outward

Scaleway grew out of Online SAS, a French hosting company founded in 1999 that became known for its Dedibox dedicated servers. The Scaleway brand arrived in 2015 with an unusual bet on ARM-based cloud servers, and over the following decade the company built out a broad public cloud while keeping the dedicated-server heritage alive. Today the platform is organised around a console at console.scaleway.com, a unified API, and a hierarchy of Organizations and Projects that every resource belongs to.

The catalog is wider than most European providers of comparable size. On the compute side there are Instances (virtual machines), Elastic Metal (bare metal billed by the hour or month), Apple silicon Mac mini servers, and GPU Instances. On the container side there are Kubernetes Kapsule, Kubernetes Kosmos, Container Registry, Serverless Containers, Serverless Functions and Serverless Jobs.

Storage covers Object Storage, Block Storage and a Serverless SQL Database, while Managed Database for PostgreSQL and MySQL and Managed Database for Redis handle relational and in-memory needs. Networking includes VPC, Private Networks, Public Gateways, Load Balancers, Domains and DNS, and Edge Services. Around the edges sit Secret Manager, Cockpit for observability, Transactional Email, IAM, and a family of AI products including Generative APIs and Managed Inference.

That breadth is why people buy Scaleway cloud account access rather than a plain VPS: the same login that runs a two-euro Instance can run a Kubernetes cluster, a GPU node and an S3-compatible bucket, all in the same Project and on the same invoice. For an orientation to how cloud accounts are structured in general, the Journal explainer on tenants, projects and subscriptions maps Scaleway's Organization and Project model onto the equivalents at AWS, Azure and Google Cloud.

Geographically, Scaleway operates three regions: Paris (fr-par), Amsterdam (nl-ams) and Warsaw (pl-waw), each with multiple availability zones. Every zone is within the European Union, which is the basis of the sovereignty argument treated in its own section below. A Scaleway virtual machine account therefore never has to answer the question of where the data is; it is in France, the Netherlands or Poland, and you chose which.

Finally, note the Dedibox line. Dedicated servers under the Dedibox name are managed through a separate console inherited from Online SAS and are distinct from Elastic Metal. Both give you physical hardware, but Elastic Metal is integrated into the Scaleway console, API and Private Networks, while Dedibox is the classic monthly dedicated-server product. This page focuses on the Scaleway console side.

The toolset behind the console

Rather than repeat the catalog, this section lists the capabilities that will actually shape your first month. Each is available to any Scaleway account for developers and can be driven from the console, the API, the scw CLI or the Terraform provider.

• Launch Instances from DEV1, PLAY2, PRO2, POP2, ENT1 and COPARM1 families with a choice of Ubuntu, Debian, CentOS-derived, Fedora and other images.
• Create Kapsule clusters with autoscaling node pools, or Kosmos clusters that accept nodes from any provider.
• Provision Elastic Metal servers by the hour and attach them to the same Private Network as your Instances.
• Rent Mac mini servers with Apple silicon for iOS and macOS build pipelines.
• Reserve GPU Instances built on NVIDIA H100, L4 and L40S hardware for training and inference.
• Store objects in S3-compatible buckets with Standard, One Zone - IA and Glacier storage classes and lifecycle rules.
• Attach Block Storage volumes with selectable IOPS tiers and snapshot them independently of the Instance.
• Deploy Serverless Functions in Node.js, Python, Go, PHP or Rust, run Serverless Containers from any image, and schedule Serverless Jobs.
• Run PostgreSQL, MySQL and Redis as managed services with automated backups and read replicas where supported.
• Build a VPC with Private Networks, route traffic through a Public Gateway, and expose services through a managed Load Balancer.
• Control access with IAM users, applications, groups and policies, and issue API keys scoped to a Project.
• Observe everything through Cockpit, which exposes metrics and logs in a hosted Grafana, and set billing alerts in the console.

Two items stand out for anyone deciding whether to buy Scaleway VPS account access as opposed to a bigger platform. The first is Kosmos, the multi-cloud Kubernetes product, which lets you keep a Scaleway control plane while attaching worker nodes from Hetzner, a home lab or another cloud. The second is Transactional Email, a managed SMTP and API service that removes the need to fight outbound port restrictions on Instances for application mail.

Also notice the Serverless SQL Database, a PostgreSQL-compatible service billed on consumption, which pairs naturally with Serverless Functions for side projects. The Journal piece on serverless versus VPS for side projects uses this kind of stack as a worked example.

One variant, one meaning: Free Trial with $100 Credit

This listing offers a single variant, and its name is a complete description. We do not extend the name with features, limits or promises, and the support channels are the right place for any question the name does not answer. Anyone who set out to buy verified Scaleway account access will receive exactly this account: one that has completed Scaleway's ordinary sign-up and is active, with no additional verification performed by AllCloudShop.

Free Trial with $100 Credit

The Free Trial with $100 Credit is a Scaleway account in trial status that carries a credit balance of $100. Two facts are stated: the account is on a free trial, and the credit balance is $100. Both are taken at face value. The variant does not name a compute quota, a region, a service entitlement, a validity period for the balance, or an upgrade path, so treat every one of those as a question for Telegram or WhatsApp rather than an assumption. It is priced at $45.

Who is it for? Anyone who wants to evaluate Scaleway with real spend behind the evaluation: a developer trying Kapsule for the first time, a startup comparing European providers before committing, a data team checking whether a GPU Instance type is available and performs as expected, or an agency building a proof of concept for a client that requires EU-hosted infrastructure. Because the balance is drawn down against ordinary usage, the trial behaves like a normal account with a head start, which is a more informative test than a feature-limited sandbox.

If you intended to buy Scaleway account with credits at a larger balance, or you are looking for a Scaleway credits account without trial status, message support first. Availability of other configurations changes, and only the support channels can confirm what is on hand.

Where Scaleway earns its place

Scaleway fits best where two things coincide: a requirement for European hosting and a desire for managed building blocks rather than raw servers alone. The profiles below are the ones that most often buy Scaleway cloud account access through this page, and each has its own reason for treating a Scaleway account for sale as the starting point.

Developers

A Scaleway account for developers is a compact playground with an unusually full toolbox. A DEV1 Instance for a side project, a Serverless Function for a webhook, a bucket for uploads and a Serverless SQL Database can all be stood up in an afternoon with scw. The Terraform provider is mature enough to describe the whole thing in a few dozen lines. If your work involves iOS, the Mac mini servers remove the need to keep a physical build machine under a desk.

Businesses with data residency obligations

A Scaleway account for business often begins with a compliance conversation. A company that handles personal data of EU residents, or that sells to public-sector clients in France or Germany, may need to show that infrastructure, provider and legal jurisdiction are all inside the Union. Scaleway satisfies that on every count, and Managed Database, Object Storage and Load Balancer mean the business does not have to self-host the pieces that are easy to get wrong. The Journal guide to data residency and sovereignty explains what the obligations actually require.

Startups

A Scaleway account for startups usually centres on Kapsule. A managed control plane, autoscaling node pools and a container registry in the same console let a small team run a production cluster without a dedicated platform engineer. Costs stay predictable because Instances are billed hourly with monthly caps, and the credit balance on this variant is enough to prove the architecture before the first real invoice. The Journal article on choosing a cloud provider for a startup discusses when a European provider is the right first home.

Agencies and consultancies

Agencies value the Organization and Project split. Each client gets a Project, each Project gets its own IAM policies and API keys, and the invoice can be read per Project. That structure keeps a Scaleway account for cloud computing work across many clients from turning into a tangle.

AI teams and researchers

A Scaleway account for AI development is one of the platform's strongest cases. GPU Instances on H100, L4 and L40S hardware are available by the hour, Object Storage holds datasets and checkpoints, and Managed Inference and Generative APIs give access to hosted models without running any GPU at all.

A team can fine-tune on an H100 Instance, store the result in a bucket, and serve it from a smaller L4 Instance behind a Load Balancer, all inside one Project. Anyone considering a Scaleway GPU account should read the Journal explainers on GPU cloud instances and running LLM inference on cloud GPUs to size the work correctly.

The common thread is that Scaleway rewards teams who want managed services without leaving Europe. A team that only wants the cheapest possible virtual machine and does not care where it runs has other options, some of which are compared later on this page.

Configuring the account: Organization, Projects, IAM, scw and a first Instance

This is the sequence to follow once the credentials arrive. It is written for someone who has just decided to buy Scaleway account online and wants the account in a professional state before any workload runs.

Secure the Organization

Log in at the console, change the password, and enable multi-factor authentication under your user profile. Then open the Organization settings and confirm the Organization name and contact details. The Organization is the top of the hierarchy; everything else lives beneath it. Read the Journal guide on multi-factor authentication for cloud accounts if you want the reasoning behind each step.

Check billing and the credit balance

Open the Billing section. You should see the credit balance the variant name describes, the payment methods on file, and the consumption view. Create a billing alert straight away: Scaleway lets you define a threshold in the console and receive a notification when consumption reaches a percentage of it. That single setting is the difference between discovering a forgotten GPU Instance in a week and discovering it in a month. The Journal guide to cost monitoring and budget alerts covers what to set the threshold to.

Create Projects

A default Project exists already. Create additional Projects for each environment or client. Resources cannot be moved between Projects afterwards, so decide the split before creating anything. Naming them by purpose, such as prod, staging and sandbox, will save confusion later.

Set up IAM and API keys

Scaleway's IAM model has users, applications, groups and policies. Policies grant permission sets, such as InstancesFullAccess or ObjectStorageReadOnly, at Organization or Project scope. Create an application for each automated system, attach a policy limited to the right Project, and generate an API key for it. An API key is an access key plus a secret key; the secret is shown once. Store both in a password manager or a secrets tool. Scaleway's IAM documentation lists every permission set, and the Journal article on managing API keys and secrets gives a sensible rotation routine.

Install and initialise scw

The scw CLI is a single binary for Linux, macOS and Windows. Run scw init, paste the access key and secret key, choose a default Organization, Project, region and zone, and the tool writes a configuration file. From then on scw instance server-type list zone=fr-par-1 shows what is available and scw account project list confirms which Projects the key can see. The Scaleway CLI documentation covers profiles for switching between Projects.

Add an SSH key and launch the first Instance

Upload your public key under Project settings so that every new Instance receives it. Then create the first machine: scw instance server create type=DEV1-S image=ubuntu_jammy zone=fr-par-1 name=web-1 ip=new. The Instance boots within a minute or two with a flexible public IP attached. Log in, create a non-root user, disable password authentication and run the first-hour security checklist.

Tighten the security group

Every Instance belongs to a security group, which is Scaleway's stateful firewall. The default group allows inbound traffic broadly, so create a new group that permits only 22, 80 and 443 inbound, set it as the Project default, and assign it to the Instance. Outbound SMTP ports are disabled on Instances by default; if your application sends email, use Transactional Email or a third-party API rather than fighting that restriction.

Understand quotas

Each Organization carries quotas per product and zone, visible in the console. The variant name on this listing does not describe any quota, so check the console for the figures that apply to your account. When you need more, the console has a request path. The Journal piece on service quotas and vCPU ceilings explains why providers start conservatively.

The European sovereignty argument, without the slogans

"Sovereign cloud" has become a marketing phrase, so it is worth being precise about what Scaleway actually offers. Three things are true and verifiable. The provider is a French company within the Iliad Group. Every region and availability zone is within the European Union. And the contractual relationship, the support organisation and the data-processing terms are all European. What that means for a given customer depends on their own obligations, and no provider can make a compliance decision for you. That is the honest frame for weighing any Scaleway account for sale: geography and corporate domicile are settled, and the rest is your design.

Regions and zones behave differently for different products. Instances, Block Storage and Kapsule are zonal: you pick a zone such as fr-par-2 and the resource lives there. Object Storage, Load Balancer and several managed services are regional and span the zones of a region. Not every Instance type is offered in every zone, and GPU Instances in particular tend to appear in Paris first, so query availability with scw before planning. The Journal introduction to regions and availability zones gives the general model.

For resilience, the sensible pattern is two zones in one region: Instances or Kapsule node pools in fr-par-1 and fr-par-2, a regional Load Balancer in front, and Object Storage for shared state. Cross-region designs, such as Paris plus Warsaw, add latency and complexity that few workloads at this scale need. If your obligation is simply "data stays in the EU", any of the three regions satisfies it; if it is "data stays in France", only fr-par does.

One honest caveat. A European provider does not by itself make a workload compliant. Encryption, access control, logging and contracts matter just as much to a Scaleway account for business use, and the Journal piece on cloud compliance basics lists what auditors actually ask for. Scaleway gives you the geographic and corporate half of the answer; the operational half is still yours.

Instances, Elastic Metal, Apple silicon and GPU: four ways to get compute

Compute on Scaleway comes in four distinct shapes, and choosing well means matching the shape to the workload rather than reaching for the familiar virtual machine every time.

Instances

Instances are the virtual machines. Families are named by intent: DEV1 for development, PLAY2 for light general use, PRO2 for production with a balanced profile, POP2 for performance-oriented work with memory- and compute-optimised subtypes, ENT1 for enterprise workloads on dedicated resources, and COPARM1 for ARM-based servers on Ampere processors. Each Instance gets a flexible IP that you can detach and move, a security group, local or Block Storage, and optional membership of Private Networks. Billing is hourly with a monthly cap. The Instances documentation describes each family and the zones where it is offered.

Elastic Metal

Elastic Metal is bare metal with cloud manners. You order a physical server, it is provisioned automatically, you can bill it hourly or monthly, and it joins Private Networks like any Instance. Use it for databases that need every cycle, for licensing that counts sockets, or for workloads that misbehave under virtualisation. The Journal comparison of dedicated servers versus cloud VMs gives the decision criteria, and they apply directly here.

Apple silicon

Mac mini servers with Apple silicon are rented by the hour with a minimum commitment. They exist for one job above all: building and testing macOS, iOS, iPadOS and watchOS applications in the cloud rather than on a laptop or a machine in a cupboard. Access is over VNC and SSH, and the servers can be attached to Private Networks.

GPU Instances

GPU Instances are where a Scaleway GPU account earns its keep. The current generation is built on NVIDIA H100, L4 and L40S accelerators in single- and multi-GPU configurations, with images that ship the NVIDIA driver and container toolkit preinstalled. They are billed hourly, which suits the fine-tuning runs and batch inference that a Scaleway account for AI development typically exists for, far better than a monthly reservation would. The Journal article on training versus inference infrastructure helps decide which GPU type belongs at which stage.

A last practical note: if you buy Scaleway VPS account access with the expectation of a single flat monthly bill, remember that every shape here is metered. The monthly cap on Instances gives predictability, but a fleet of GPU Instances left running has no such ceiling. That is precisely why the billing alert in the setup section matters.

Kapsule and Kosmos: managed Kubernetes the Scaleway way

Scaleway offers two managed Kubernetes products, and the distinction is one of the most useful things to understand before creating a Scaleway Kubernetes account workload.

Kapsule is the conventional product. Scaleway hosts and operates the control plane, and worker nodes are Scaleway Instances grouped into node pools. Each pool has an Instance type, a zone, a size range and optional autoscaling and autohealing. You can mix pools, for example a PRO2 pool for general workloads and an L4 GPU pool for inference, and let the cluster autoscaler add or remove nodes as demand changes. The cluster comes with the Scaleway cloud controller manager and CSI driver, so a Kubernetes Service of type LoadBalancer creates a Scaleway Load Balancer and a PersistentVolumeClaim provisions Block Storage without extra setup.

Kosmos is the multi-cloud variant. The control plane is still on Scaleway, but nodes can be any Linux machine that can reach it: an Instance in another region, an Elastic Metal server, a machine at a different provider, or hardware in your own office. This is unusual among managed Kubernetes products and makes Kosmos a practical way to unify a fleet that already spans providers, or to keep the control plane in the EU while placing some nodes elsewhere for latency reasons.

Both products expose a standard kubeconfig, support the usual add-ons, and can be created from the console, the API, scw or Terraform. A minimal cluster with scw k8s cluster create name=demo version=latest pools.0.name=default pools.0.node-type=PRO2-XXS pools.0.size=2 is a one-line experiment that the credit on this variant comfortably covers for a trial period. The Kubernetes documentation explains version lifecycle, upgrades and the add-on catalog.

Container Registry sits beside both. It is a private, regional registry for your images with a per-namespace access model, and it integrates with IAM so that a Kapsule cluster can pull images using a Project-scoped API key. The Journal article on container registries describes how to structure namespaces, and the piece on managed Kubernetes for small teams is a fair test of whether a Scaleway Kubernetes account is the right amount of infrastructure for your stage.

Two cautions. First, a control plane on the free tier of Kapsule may carry different availability characteristics from the paid dedicated control plane options; read the current documentation before running production on the cheapest option. Second, the cost of a cluster is dominated by its nodes and Load Balancers, not by the control plane, so size pools deliberately and let autoscaling shrink them at night. Neither caution is peculiar to a Scaleway cloud account for sale through this listing; they apply to every Kapsule cluster.

Object Storage, Block Storage and the Serverless family

Object Storage

Scaleway Object Storage is an S3-compatible service organised into regional buckets. Three storage classes cover the usual lifecycle: Standard for active data, One Zone - IA for infrequently accessed data that can live in a single zone, and Glacier for archives that tolerate a restore delay. Lifecycle rules move objects between classes automatically. Bucket policies, IAM permission sets and pre-signed URLs handle access.

Because the API is S3-compatible, tools such as the AWS CLI, rclone, restic and any S3 SDK work with an endpoint change. Scaleway's Object Storage documentation lists the endpoints per region and the supported S3 operations, and the Journal explainer on object storage across providers shows where Scaleway's implementation sits among S3-compatible options.

Block Storage

Block Storage volumes are network-attached disks with selectable performance tiers, attachable to Instances and to Kapsule nodes through the CSI driver. Volumes persist independently of the Instance, can be snapshotted, and can be resized. Use them for databases, for anything that must survive an Instance being replaced, and for data you want to move between machines. The Journal comparison of block versus object storage is the right reference when deciding which layer a dataset belongs on.

Serverless Functions, Containers and Jobs

The Serverless family is broader than at most providers of Scaleway's size. Serverless Functions run code in supported runtimes in response to HTTP requests, cron schedules or messages, and scale to zero when idle. Serverless Containers do the same for any container image, which removes runtime restrictions. Serverless Jobs run batch tasks to completion on a schedule or on demand, ideal for nightly processing, report generation or data pipelines that do not warrant a permanently running Instance. All three bill on consumption and integrate with Cockpit for logs and metrics.

Managed databases and Serverless SQL

Managed Database for PostgreSQL and MySQL provides a provisioned database Instance with automated backups, optional high availability and read replicas. Managed Database for Redis covers caching and queues. Serverless SQL Database is a newer PostgreSQL-compatible option billed on actual usage that suits intermittent workloads. The Journal piece on managed databases versus self-hosted helps decide when the managed price is worth paying, and the guide to backups and point-in-time recovery explains what "automated backups" should mean before you rely on it.

Put together, these services are what make a Scaleway virtual machine account feel like a full cloud rather than a hosting panel. A web application can run on Instances or Kapsule, store files in a bucket, keep its database managed, offload background work to Serverless Jobs and send mail through Transactional Email, all without a single self-hosted supporting service.

How the $100 credit is consumed, and what happens when it is gone

Scaleway bills most resources hourly and invoices monthly. A credit balance on the account is applied against the invoice before any payment method is charged, so the first invoices on the Free Trial with $100 Credit variant will draw on the balance until it is exhausted. The console's consumption view shows both the running usage for the current period and the remaining credit, and the billing alert described in the setup section can be pointed at either.

What a credit balance does not change is the meter. A GPU Instance costs the same per hour whether it is funded by credit or by card, and a forgotten Instance consumes the balance just as readily as a useful one. Several Scaleway services carry a monthly free allowance in addition to any credit, notably Object Storage, Serverless Functions and Serverless Containers, and those allowances are applied before either credit or payment.

The Journal explainer on how promotional credits work describes the general mechanics and the questions to ask about expiry, and the piece on free trial versus pay-as-you-go covers what typically changes when a trial period ends.

Because the variant name states only the trial status and the balance, we do not state an expiry date, a spending limit, or the conditions under which Scaleway moves an account from trial to full status. Those details are Scaleway's to define and may change. Read the current terms in the console, and if you need a particular answer before ordering, ask support. That is also the channel for anyone who wants to buy Scaleway account with credits beyond this balance, or who is looking for a Scaleway credits account in a non-trial state.

How ordering through AllCloudShop works

Choose the variant, complete checkout, and the team prepares the account credentials for delivery through the channel selected at purchase. What arrives is a Scaleway login in an active state with the credit balance the name describes, which is all that the phrase buy verified Scaleway account has ever meant on this page. From that moment the account is operated by you under Scaleway's terms of service, which is why the first actions must be changing the password and enabling multi-factor authentication.

Buyers who want to buy Scaleway account online in quantity, or who need several Scaleway accounts for sale delivered together, should ask about availability first, since it varies. Post-delivery support covers the login and the meaning of the variant name; operating your Instances, clusters and buckets is your responsibility, and the documentation links throughout this page are the best starting point.

Seven traps for a new Scaleway account

These come up repeatedly with people who buy Scaleway account with credits and start deploying the same afternoon. None is hard to avoid once named.

1. Creating resources in the wrong Project. Resources cannot be moved between Projects. Set the default Project in scw and check the Project selector in the console before every create action until it is habit.
2. Leaving the default security group open. The default group is permissive. Create a restrictive group, make it the Project default, and assign it before exposing anything. The Journal article on cloud firewalls and security groups explains stateful rules if the concept is new.
3. Expecting outbound email to just work. SMTP ports on Instances are restricted by default. Use Transactional Email or an external API for application mail rather than a local mail server. The port 25 explainer covers why providers do this.
4. Mixing up zonal and regional resources. An Instance in fr-par-1 cannot attach a Block Storage volume created in fr-par-2, but it can reach a regional bucket from either zone. Check the scope of each product before designing around it.
5. Running GPU Instances without a stop routine. Hourly GPU billing is only a bargain if the Instance is stopped when idle. Script a shutdown, use Serverless Jobs for batch work, or at least set a billing alert.
6. Sharing one API key across every system. Create an IAM application per system with a Project-scoped policy. A leaked CI key should not be able to delete production. The Journal primer on cloud IAM fundamentals is short and worth the ten minutes.
7. Skipping backups because a service is managed. Managed databases have automated backups, but Instances do not back themselves up. Snapshot Block Storage volumes, export database dumps to Object Storage, and test a restore. The guide to cloud backup strategies describes a routine that fits this platform.

An eighth, softer trap is treating a Scaleway account for cloud computing as if it were a hyperscaler account. The catalog is broad, but it is not infinite, and a design that assumes a specific managed service exists should be checked against the console before code is written.

Scaleway next to OVHcloud, Hetzner, DigitalOcean and Civo

Buyers evaluating a Scaleway cloud account for sale are almost always comparing it with a few named alternatives. The table summarises the pairings; the linked Cloud Matrix pages go into depth.

Read across a row and the pattern is clear: Scaleway is the European provider with the most complete managed catalog at its price point, and its trade-off is a footprint confined to three EU regions. If that footprint is a requirement rather than a limitation, the comparison usually ends in Scaleway's favour, and a Scaleway account for startups whose customers are mostly European sits squarely in that case.

If you need North American or Asian presence, the Hetzner account listing or the OVHcloud account listing may fit better, and teams whose whole world is Kubernetes should look at the Civo account listing. The broader Journal essay on hyperscalers versus developer clouds places Scaleway near the middle of that spectrum, which is exactly where its catalog suggests.

Whether you buy Scaleway cloud account access from this listing or arrive at Scaleway another way, the decision framework is the same: European jurisdiction, managed services, hourly billing, and a catalog that is broad enough for most products and honest about its limits.

Scaleway account FAQ