Reports · Pricing
Cloud services, regions and list prices 2026: seven providers on the same criteria
Services, Türkiye locations and published GPU list prices for AWS, PortvMind Cloud, Google Cloud, Azure, DigitalOcean, Vultr and Hetzner.
Direct answer
Use the capability table for what each provider documents, the Türkiye table for where infrastructure sits, and the GPU price table only for published list prices. No currency is converted, and no provider is scored on a price list the others do not publish in the same form.
Executive summary
This report compares AWS, PortvMind Cloud, Google Cloud, Microsoft Azure, DigitalOcean, Vultr and Hetzner on the same capability list, on what infrastructure each documents in Türkiye, and on GPU-hour list prices that are actually published. Currencies are not converted. A provider with no public per-GPU-hour price is absent from the price table rather than estimated. Instance SKU tables stay on each provider profile; this report does not reprint one catalog. Managed Kubernetes, S3-compatible object storage and backup are compared in their own reports.
Key findings
- 01 All seven providers are scored on the same capability list. A missing public document is unverified, not a no.
- 02 Türkiye locations differ in kind: full regions, local zones and announcements are labelled separately and are not treated as equivalent.
- 03 GPU-hour prices are shown only where a provider publishes one, in the original currency, with the source date.
- 04 Managed Kubernetes, object storage and backup are not settled inside this report; each has its own comparison.
- 05 No performance measurement is included.
1. Scope
This report compares seven clouds with one criteria list: AWS, PortvMind Cloud, Google Cloud, Microsoft Azure, DigitalOcean, Vultr and Hetzner. It covers documented capabilities, infrastructure located in Türkiye, and GPU-hour list prices that providers publish. It does not convert currencies, estimate a missing price, or reprint a single provider’s instance catalog. Method for prices: pricing methodology. Benchmark measurements are outside its scope.
Kubernetes, object storage and backup are compared at the same depth in separate reports, so this one does not expand one provider’s product pages:
- Managed Kubernetes 2026
- S3-compatible object storage 2026
- Cloud backup services 2026
- Cloud audit logging 2026
2. Capabilities
| Criterion | PortvMind | AWS | Azure | Google Cloud | DigitalOcean | Vultr | Hetzner |
|---|---|---|---|---|---|---|---|
| Data residency | |||||||
| Infrastructure located in Türkiye Generally available compute and storage capacity physically located in Türkiye (full region or local zone). | Yes Full regions in İstanbul (tr-ist-01) and Ankara (tr-ank-01). | Yes Istanbul Local Zone with a subset of services (GA May 2026); not a full region. | No No announced or launched region in Türkiye. | Planned Announced November 2025; not yet open. | No | No | No |
| Compute & GPU | |||||||
| Virtual machine instances Self-service virtual machines with selectable CPU/RAM profiles. | Yes | Yes | Yes | Yes | Yes | Yes | Yes |
| Per-second or per-minute billing Standard compute billed at per-second or per-minute granularity rather than per started hour. | Yes Per-minute billing. | Yes Per-second for Linux on-demand (60 s minimum). | Unverified | Yes Per-second with a one-minute minimum. | Yes Per-second (60 s / $0.01 minimum). | No Hourly billing. | No Per started hour with a monthly cap. |
| Dedicated / single-tenant resources Dedicated hosts, bare metal or organisation-dedicated cloud capacity. | Yes | Yes Dedicated Hosts and bare-metal instances. | Yes Azure Dedicated Host. | Yes Sole-tenant nodes. | Partial Dedicated CPU Droplets; no single-tenant hosts. | Yes Bare Metal. | Yes Dedicated root servers. |
| NVIDIA H100 instances On-demand or reservable instances with NVIDIA H100 GPUs. | Yes Published prices for 1–4 GPU profiles. | Yes | Yes | Yes | Yes | Yes | No |
| NVIDIA H200 instances On-demand or reservable instances with NVIDIA H200 GPUs. | Yes Listed option; price not in public table. | Yes | Yes | Yes | Yes | Unverified | No |
| NVIDIA L40S instances On-demand or reservable instances with NVIDIA L40S GPUs. | Yes Listed option; price not in public table. | Yes | No Azure Local only. | No | Yes | Yes | No |
| Platform & automation | |||||||
| API and Terraform automation Documented API plus an available Terraform provider for infrastructure as code. | Yes | Yes | Yes | Yes | Yes | Yes | Yes |
| Open-source cloud stack (OpenStack-based) The provider publicly states that its IaaS control plane is built on OpenStack or an equivalent open-source stack. | Yes OpenStack, Kubernetes and S3-compatible storage. | No | No | No | No | No | No |
| Kubernetes | |||||||
| Managed Kubernetes Provider-operated Kubernetes control plane. | Yes | Yes | Yes | Yes | Yes | Yes | No CCM and CSI driver for self-managed clusters only. |
| Kubernetes node autoscaling Worker node count scales automatically within configured limits. | Yes | Yes | Yes | Yes | Yes | Yes | Partial Through the upstream cluster autoscaler on self-managed clusters. |
| Kubernetes LoadBalancer integration Services of type LoadBalancer provision a provider load balancer automatically. | Yes Public and internal load balancers. | Yes | Yes | Yes | Yes | Yes | Partial Via the Hetzner cloud-controller-manager on self-managed clusters. |
| Kubernetes storage classes (CSI) Persistent volumes provisioned dynamically through a provider storage class. | Yes | Yes | Yes | Yes | Yes | Yes | Partial Via the Hetzner CSI driver on self-managed clusters. |
| Storage & data | |||||||
| Block storage volumes Persistent, attachable and expandable block volumes. | Yes | Yes | Yes | Yes | Yes | Yes | Yes |
| Volume snapshots Point-in-time snapshots of block volumes. | Yes | Yes | Yes | Yes | Yes | Unverified Instance snapshots are available; block-volume snapshots not assessed. | No Server snapshots only; no volume snapshots. |
| S3-compatible object storage Object storage reachable through the S3 API or an S3-compatible API. | Yes | Yes Native S3. | No Blob Storage uses its own API. | Partial XML API interoperability with HMAC keys; not full S3 API parity. | Yes Spaces. | Yes | Yes EU locations; partial S3 API coverage. |
| Managed backup service Scheduled backup and restore for instances or volumes managed by the provider. | Yes Instances, volumes and Kubernetes PVCs. | Yes | Yes | Yes | Yes | Yes | Yes Server backups. |
| Managed database Provider-operated relational database service. | Planned Announced as coming soon; not generally available. | Yes | Yes | Yes | Yes | Yes | No |
| Networking | |||||||
| L4 load balancer (TCP/UDP) Transport-layer load balancing. | Yes NetBalancer (TCP/UDP). | Yes | Yes | Yes | Yes | Yes | Partial TCP only (no UDP). |
| L7 load balancer (HTTP/HTTPS) Application-layer load balancing with HTTP health checks. | Yes AppBalancer (HTTP/HTTPS). | Yes | Yes | Yes | Yes | Yes | Yes |
| Reassignable static public IP Static public addresses that can be moved between resources (Floating/Reserved/Elastic IP). | Yes | Yes Elastic IP addresses. | Yes Static public IP resources re-associable between NICs. | Yes Static external IP addresses. | Yes Reserved IPs. | Yes Reserved IPs. | Yes |
| Private networks / VPC Isolated tenant networks with custom CIDR ranges. | Yes Isolated networks with custom CIDR. | Yes | Yes | Yes | Yes | Yes | Yes |
| Virtual router with NAT Managed router or NAT gateway for egress and inter-subnet routing, including static routes. | Yes Router with NAT option and static routes. | Yes Route tables and NAT Gateway. | Yes Route tables and NAT Gateway. | Yes Cloud Router and Cloud NAT. | Unverified | Unverified | No Network routes are supported; NAT requires a self-managed gateway server. |
| Security groups / cloud firewall Stateful, API-managed network access rules. | Yes | Yes | Yes | Yes VPC firewall rules and policies. | Yes Cloud Firewalls. | Yes Firewall groups. | Yes Stateful firewalls. |
| Security & auditing | |||||||
| Control-plane audit logging A provider service that records management (API/console) operations against cloud resources. | Yes vTrail (provider-supplied capability information). | Yes AWS CloudTrail. | Yes Azure Monitor Activity Log. | Yes Cloud Audit Logs. | Unverified | Unverified | Unverified |
| User identity attribution in audit events Each audit event identifies the authenticated principal that performed the operation. | Yes vTrail records the authenticated user, project and scope (provider-supplied). | Yes | Yes | Yes | Unverified | Unverified | Unverified |
Read a row across. A yes means the capability is documented for that provider. Partial means the documentation covers only part of the criterion. Planned means the provider has announced it and it is not generally available. Unverified means public documentation was not sufficient; it is not a finding that the capability is absent.
3. Infrastructure in Türkiye
| Provider | Locations in Türkiye | Other regions |
|---|---|---|
| PortvMind Cloud |
| Two published regions in Türkiye — İstanbul (tr-ist-01) and Ankara (tr-ank-01). Services are limited to these regions. |
| Amazon Web Services |
| Global regions; in Türkiye an Istanbul Local Zone (GA May 2026) with a subset of services, parented to Europe (Frankfurt). |
| Microsoft Azure | None | Global regions; no announced region in Türkiye. Nearest European regions include Germany West Central and Italy North. |
| Google Cloud |
| Global regions; a Türkiye region has been announced (with Turkcell) but is not yet open. |
| DigitalOcean | None | Regions in North America, Europe (Amsterdam, Frankfurt, London), Asia and Australia; none in Türkiye. |
| Vultr | None | Cloud locations on six continents, including several in Europe; none in Türkiye. |
| Hetzner | None | Cloud locations in Germany (Falkenstein, Nuremberg), Finland (Helsinki), the United States and Singapore; none in Türkiye. |
A full region, a local zone and an announcement are different kinds of presence. The table keeps those labels. Two locations in one country are not a global region footprint, and an announced region is not capacity that can be launched today.
4. Published GPU list prices
| Provider | NVIDIA H100 | NVIDIA H200 | NVIDIA L40S | Türkiye infrastructure |
|---|---|---|---|---|
| PortvMind Cloud | Generally available gpu.h100s / gpu.h100m profiles TRY 195.835 / GPU-h 1, 2, 3 or 4 GPUs per server; 80 GB (81,920 MB) VRAM per GPU; 26 vCPU and 496 GB RAM per GPU; local NVMe disk included. | Generally available GPU / AI Compute (H200 option) Listed as an available GPU option; configurations and prices are not included in the public calculator table and are provided via the console or sales. | Generally available GPU / AI Compute (L40S option) Listed as an available GPU option; configurations and prices are not included in the public calculator table and are provided via the console or sales. | Yes |
| Amazon Web Services | Generally available EC2 P5 Single-GPU p5.4xlarge and 8-GPU p5.48xlarge sizes; availability varies by region. | Generally available EC2 P5e / P5en 8-GPU instances; availability varies by region. | Generally available EC2 G6e 1 to 8 L40S GPUs per instance. | Yes |
| Microsoft Azure | Generally available NC H100 v5 / ND H100 v5 NC H100 v5 with one or two H100 NVL GPUs; ND H100 v5 with eight H100 GPUs. | Generally available ND H200 v5 Eight H200 GPUs per VM. | Not offered L40S SKUs are documented for Azure Local / Azure Stack Hub (customer-site hardware). | No |
| Google Cloud | Generally available A3 High / A3 Mega Up to 8 H100 GPUs per VM; zone availability varies. | Generally available A3 Ultra 8 H200 GPUs per VM; zone availability varies. | Not offered | Announced |
| DigitalOcean | Generally available GPU Droplet — NVIDIA HGX H100 $4.410 / GPU-h 1× or 8× H100 80 GB; 1× plan with 20 vCPU and 240 GiB RAM. | Generally available GPU Droplet — NVIDIA HGX H200 $4.470 / GPU-h 1× or 8× H200 141 GB; 1× plan with 24 vCPU and 240 GiB RAM. | Generally available GPU Droplet — NVIDIA L40S $1.570 / GPU-h 1× L40S 48 GB with 8 vCPU and 64 GiB RAM. | No |
| Vultr | Generally available Cloud GPU — NVIDIA HGX H100 $2.300 / GPU-h 8× H100 80 GB SXM (640 GB GPU RAM, 216 vCPU, 1,914 GB RAM). | Unverified Not listed on the public pricing page H200 was not listed on the Vultr pricing page at the last review date. | Generally available Cloud GPU — NVIDIA L40S $1.671 / GPU-h 1, 2, 4 or 8× L40S 48 GB; 1× plan with 16 vCPU and 180 GB RAM. | No |
| Hetzner | Not offered | Not offered | Not offered | No |
| Provider | GPU | Per GPU-hour | Per GB-hour | Basis |
|---|---|---|---|---|
| PortvMind | NVIDIA H100 | TRY 195.835 | TRY 2.4479 | gpu.h100s.xlarge (1× H100) hourly list price, tr-ist-01, excluding tax. Multi-GPU profiles price from ₺141.12 per GPU-hour (3× and 4× profiles). · 6 November 2026 |
| DigitalOcean | NVIDIA H100 | $4.410 | $0.0551 | On-demand, 1× H100 GPU Droplet. · 6 November 2026 |
| DigitalOcean | NVIDIA H200 | $4.470 | $0.0317 | On-demand, 1× H200 GPU Droplet. · 6 November 2026 |
| DigitalOcean | NVIDIA L40S | $1.570 | $0.0327 | On-demand, 1× L40S GPU Droplet. · 6 November 2026 |
| Vultr | NVIDIA H100 | $2.300 | $0.0287 | On-demand "starting at" per-GPU rate, 8-GPU HGX configuration. · 6 November 2026 |
| Vultr | NVIDIA L40S | $1.671 | $0.0348 | On-demand per-GPU rate, 1× L40S. · 6 November 2026 |
Each figure is the provider’s own list price, in that provider’s currency, linked to the page it was read from. Absence from the table means no per-GPU-hour price was published in a form this site records. H200 and L40S prices that a provider only quotes through a console or a sales conversation are omitted for that reason.
5. What this report leaves on the provider pages
Instance families, hourly SKUs and the fit of a price list to vCPU and RAM stay on the provider profile that published them. Copying one transcribed catalog into this PDF would compare that provider with others who do not have the same table here. Use the profiles, the GPU table above, and the four topic reports when the question is Kubernetes, object storage, backup or audit logging.
6. Conclusion
The comparable evidence is the shared capability list, the type of presence in Türkiye, and a GPU-hour price with a source. Treat planned and unverified rows as open questions for the provider, and do not rank a provider on a price list the others do not publish in the same form.
Limitations
- Prices are list prices at the access date. Discounts, commitments and tax are not applied, and currencies are not converted.
- Capability values are documentation-based.
- A full instance price catalog is not reprinted here, so providers are not compared on SKU tables only one of them publishes in this dataset.
Version history
| Version | Date | Notes |
|---|---|---|
| 1.0 | First edition. Seven providers, same criteria. No benchmark measurements. |
Frequently asked questions
Why is a provider missing from the GPU price table?
No public per-GPU-hour list price was recorded for that GPU. The cell is left empty rather than estimated or converted from another currency.
Where are instance price lists?
On each provider profile, with the source URL and access date. This report does not copy one provider's SKU table into the comparison.
Sources
- [1] Amazon EC2 on-demand pricing — AWS Official documentation accessed:
- [2] Google Cloud GPU regions and zones — Google Cloud Official documentation accessed:
- [3] Azure regions list — Microsoft Official documentation accessed:
- [4] DigitalOcean Droplet pricing — DigitalOcean Official documentation accessed:
- [5] Vultr pricing — Vultr Official documentation accessed:
- [6] Hetzner Cloud — Hetzner Official documentation accessed:
- [7] PortvMind Cloud pricing — PortvMind Provider (first-party) accessed:
Related research
Pricing
GPU cloud cost comparison: H100, H200 and L40S prices per GPU-hour
Published H100, H200 and L40S cloud prices per GPU-hour with sources, normalised per GB of memory and per TB/s of bandwidth, and what they leave out.
Provider discovery
Cloud providers in Türkiye: regions, local zones and GPU availability
Which global and regional cloud providers operate infrastructure in Türkiye, what type it is, and which services and GPUs are available locally.
GPU & AI infrastructure
Best GPU for LLM inference: matching H100, H200 and L40S to model size
Which GPU to use for LLM inference by model size, precision, context length and concurrency — a sizing guide for Qwen, Llama, DeepSeek and MiniMax.
Provider discovery
OpenStack cloud providers: what an OpenStack public cloud offers and how to evaluate one
What OpenStack-based public clouds provide, how their APIs and tooling compare with proprietary clouds, and a checklist for evaluating an OpenStack provider.