Holori Cloud Calculator
Free, open, and account-free. The Holori Cloud Calculator is available at https://calculator.holori.com and requires no registration to use.
The Holori Cloud Calculator is a comprehensive cloud pricing comparison tool that lets you compare instance prices, storage costs, LLM API pricing, and service equivalencies across all major cloud providers — including AWS, Azure, GCP, Oracle Cloud (OCI), OVHcloud, Scaleway, and Linode. Prices are updated daily from official provider APIs.
Overview
The calculator is organized into two groups of navigation items:
Cloud Provider tabs — for browsing and comparing compute/storage pricing per provider:
- AWS
- Azure
- GCP
- More Providers (OCI, OVHcloud, Scaleway, Linode, etc.)
Feature tabs — for specialized comparison views:
- Spot Map
- Datacenter PUE
- LLM Pricing
- Compare Services
A basket is available in the top-right corner, allowing you to add instances from any provider and build a combined cost estimate.
Cloud Provider Tabs
AWS
The AWS tab displays pricing for Amazon EC2 instances. It is the default landing page of the calculator.
Filters available:
- Search Instance — search by instance name (e.g.,
c5.4xlarge) - Region — select an AWS region (e.g.,
N. Virginia (us-east-1)) - Payment Option — On-Demand, Spot, Savings Plan, or Reserved
- Operating System — Linux, Windows, etc.
- vCPUs — filter by minimum/maximum number of vCPUs
- Memory (GiB) — filter by minimum/maximum RAM
Pricing columns displayed:
- On-Demand Hourly & Monthly
- Spot price
- Savings Plan (1-year and 3-year, No Upfront)
- Reserved (1-year and 3-year, No Upfront)
Each row has an Add to Basket button to include the instance in your cost estimate. You can also click on any instance name to view a detailed page with all pricing options, instance metadata, and tags (architecture, processor type, etc.). The date of the last price retrieval from the AWS API is shown in the instance detail pop-up.
You can Export CSV to download the current filtered list.
Azure
The Azure tab works identically to the AWS tab but covers Azure Virtual Machines. The same filter options apply: region, payment option, OS, vCPUs, and memory.
Azure-specific pricing models are supported including:
- Pay-as-you-go (On-Demand equivalent)
- Spot VMs (heavily discounted preemptible instances)
- Reserved Instances (1-year and 3-year)
- Azure Hybrid Benefit (BYOL — Bring Your Own License for customers with existing Windows Server or SQL Server licenses)
For the same VM type, prices can vary by up to 300% between the cheapest and most expensive Azure region. Use the region filter to find the most cost-effective location for your workload.
GCP
The GCP tab covers Google Compute Engine instance pricing. Supported pricing models include:
- On-Demand
- Spot / Preemptible VMs
- Committed Use Discounts (1-year and 3-year)
The same filtering capabilities apply: region, OS, vCPUs, and memory.
More Providers
The More Providers section extends the comparison to smaller and European cloud providers:
- OCI (Oracle Cloud Infrastructure)
- OVHcloud
- Scaleway
- Linode (Akamai Cloud)
Each provider's tab follows the same interface as AWS, Azure, and GCP, with filters adapted to the provider's available options. Not all pricing models (e.g., spot) are available for all providers — availability depends on what each provider offers.
Feature Tabs
Spot Map
The Spot Map tab (https://calculator.holori.com/spot-capacity) is dedicated to AWS Spot Instance analysis. Spot instances use spare AWS capacity and can be up to 90% cheaper than On-Demand pricing — but AWS may reclaim them with short notice when that capacity is needed elsewhere.
This tool helps you balance cost savings with workload reliability by comparing:
Instance Type Selector — choose any EC2 instance type (e.g., c5.4xlarge) to load its metrics across all regions.
Three views:
- Pricing — spot price per region and availability zone
- Interruption Rate — historical frequency of AWS reclaiming the instance in each location
- Availability — capacity availability signals
Interruption Rate Guide:
- 🟢
< 5%— Excellent reliability, minimal interruptions - 🟡
5–15%— Moderate stability, occasional reclaims - 🔴
> 15%— Higher volatility, frequent interruptions
The cheapest region is not always the best choice. For production workloads, diversify across multiple instance types and availability zones to improve resilience while maintaining cost savings.
Data is updated every few hours to reflect the real-time nature of spot pricing.
Datacenter PUE
The Datacenter PUE tab (https://calculator.holori.com/datacenter-pue) provides environmental efficiency data for cloud datacenters across AWS, Azure, GCP, OVHcloud, and Scaleway.
This section is particularly useful for organizations with sustainability goals or regulatory requirements around carbon footprint.
What is PUE?
Power Usage Effectiveness (PUE) is the industry-standard metric for datacenter energy efficiency:
PUE = Total Facility Energy / IT Equipment Energy
A PUE of 1.0 is the theoretical perfect score (all power goes to IT equipment, none is wasted). In practice, any value below 1.2 is considered excellent.
| PUE Rating | Efficiency Level |
|---|---|
| < 1.2 | ✅ Excellent |
| 1.2 – 1.3 | 🟡 Good |
| 1.3 – 1.4 | 🟠 Average |
| > 1.4 | 🔴 Poor |
What is WUE?
Water Usage Effectiveness (WUE) measures liters of water used per kWh of IT equipment energy. Lower is better.
WUE = Annual Water Usage (L) / IT Equipment Energy (kWh)
| WUE Rating | Efficiency Level |
|---|---|
| < 0.5 L/kWh | ✅ Excellent |
| 0.5 – 0.8 L/kWh | 🟡 Good |
| 0.8 – 1.2 L/kWh | 🟠 Average |
| > 1.2 L/kWh | 🔴 Poor |
Key Statistics (as of last update)
- 104 datacenters tracked across 5 providers
- Best PUE: 1.04 — Google Cloud, Lancaster, Ohio
- Average PUE: 1.18 across all tracked facilities
- 73 datacenters rated as Eco Leaders (PUE < 1.2)
Data is sourced directly from official sustainability reports published by each provider.
LLM Pricing
The LLM Pricing tab (https://calculator.holori.com/llm) is a comprehensive AI model pricing comparison across all major LLM API providers. It covers over 1,000 models from providers including:
- OpenAI
- Anthropic
- Google (Gemini / Vertex AI)
- AWS Bedrock
- Azure AI
- Mistral
- Cohere
- DeepSeek
- xAI (Grok)
- OVHcloud
- OCI
Filters
- Search Model — search by model name or ID
- Provider — filter by one or more providers
- Category — filter by model type:
- Text Generation
- Reasoning
- Embeddings
- Code Generation
- Image Generation
- Audio / Speech
Pricing Table Columns
| Column | Description |
|---|---|
| Provider | The API provider (e.g., OpenAI, AWS Bedrock, Anthropic) |
| Model | The model name and internal identifier |
| Category | Model type (Text Generation, Reasoning, Embeddings, etc.) |
| Input / 1M tokens | Cost per million input tokens |
| Output / 1M tokens | Cost per million output tokens |
| Context | Maximum context window size |
| Features | Capability badges: Vision, Functions, JSON, Caching |
Click any column header to sort the table. Click any model name to open a detailed pricing page for that specific model.
Use the Calculator button to estimate your total cost based on your expected token usage volume.
Prices are updated regularly (last updated dates are shown at the top of the page).
Compare Services
The Compare Services tab (https://calculator.holori.com/compare) maps equivalent services across cloud providers, making it easy to find the counterpart of a familiar AWS service on Azure, GCP, or a smaller provider — or vice versa.
It currently covers 55 services across 12 categories for AWS, Azure, GCP, Oracle Cloud, OVHcloud, Scaleway, and Linode.
Categories
| # | Category | Services Covered |
|---|---|---|
| 1 | Compute | Virtual Machines, Bare Metal, GPU, Spot/Preemptible, Autoscaling, Batch Processing |
| 2 | Containers & Serverless | Kubernetes, Container Orchestration, Container Registry, Serverless Containers, Functions (FaaS), Event-Driven |
| 3 | Storage | Object Storage, Block Storage, File Storage, Archive Storage, Backup |
| 4 | Databases | Managed Relational DB, PostgreSQL, MySQL, NoSQL/Document DB, In-Memory Cache, Data Warehouse |
| 5 | Networking | Virtual Network, Load Balancer, CDN, DNS, Dedicated Connection, Firewall |
| 6 | AI & Machine Learning | ML Platform, Generative AI, AI Notebooks, Model Deployment |
| 7 | Data Analytics | Data Warehouse, ETL/Integration, Stream Processing, Search & Analytics, Monitoring |
| 8 | Security & Identity | IAM, Key Management, Secrets Management, Security Monitoring |
| 9 | Developer Tools | CI/CD, Infrastructure as Code, CLI/SDK, API Gateway |
| 10 | Management & Monitoring | Monitoring, Logging, Cost Management |
| 11 | Migration | Migration Hub, Database Migration, Data Transfer |
| 12 | Hybrid & Multi-cloud | Hybrid Platform, Multi-cloud Kubernetes, Backup & DR |
Each cell in the comparison table links directly to the official documentation page for that service. A - in a cell means the provider does not offer an equivalent service.
Data is sourced from official cloud provider documentation and was last verified in January 2025.
The Basket
The basket (accessible from the top-right corner on any page) lets you build a multi-provider cost estimate by adding instances from any provider tab.
- Click Add to Basket on any instance row to include it
- The basket persists as you navigate between provider tabs
- A total estimated monthly cost is calculated across all added items
- You can share or export your basket to collaborate with your team or present to procurement
Tips & Best Practices
Finding the cheapest region: Unlike individual provider calculators (AWS Pricing Calculator, Azure Pricing Calculator, GCP Pricing Calculator), Holori allows you to search across all regions simultaneously and sorts results from cheapest to most expensive. This is especially useful when regulatory or latency constraints don't restrict your region choice.
Comparing instance families: Use the vCPU and Memory filters to define your resource requirements and instantly see which instance types across all providers meet your needs, sorted by price.
Spot instance strategy: Use the Spot Map to identify regions where your target instance type has both a low spot price and a low interruption rate. Avoid relying solely on price when choosing spot regions for sensitive workloads.
Sustainability-driven decisions: Use the Datacenter PUE tab to choose cloud regions that align with your organization's environmental goals. For the same workload, choosing a datacenter with a PUE of 1.07 vs. 1.45 can represent a very significant reduction in wasted energy.
LLM cost planning: When evaluating AI features, use the LLM Pricing tab to compare API costs per million tokens across all major providers and models before committing to a specific API.
Go Further with Holori
The Holori Calculator is the free, open entry point into the Holori ecosystem. To go further with cloud cost management and infrastructure visibility, you can create a free account at https://app.holori.com.
The full Holori platform offers:
- Multi-provider cost dashboards — centralize costs from AWS, Azure, GCP, OCI, and more in a single view
- Automated infrastructure diagrams — connect your cloud account and auto-generate architecture diagrams
- Cost reports and filters — slice costs by team, tag, service, or region
- Budgets and alerts — set spending thresholds and get notified before you overspend
- Recommendations — receive actionable rightsizing and savings suggestions