AWS Bedrock and Azure OpenAI side by side
| AWS Bedrock | Azure OpenAI | |
|---|---|---|
| Models | Claude, OpenAI GPT-6, GPT-5.x and gpt-oss, Llama, Mistral, DeepSeek, Grok and others; Gemini not listed | OpenAI models; Claude offered with Anthropic as seller and operator |
| On-demand pricing | Per token on your AWS bill | Per token per deployment on your Azure bill |
| Reserved capacity | Provisioned Throughput per model unit per hour; no-commit, 1-month or 6-month terms | Provisioned Throughput billed hourly per PTU regardless of tokens; 1-month or 1-year reservations |
| Capacity limits | Tokens per minute per model per Region; increases are not automatic and may be denied | Tokens per minute per region, model and deployment type; increases reviewed |
| Processing in Europe | EU geographic cross-Region profiles stay in the EU; global profiles may route to any commercial Region | Data Zone EU follows the EU Data Boundary; Global deployments may process in any Azure region |
| Claude in Europe | EU cross-Region profiles | Global Standard in Sweden Central; Data Zone Standard not available in Europe |
| Retention | Data retention mode set per Region; some models keep traffic up to 30 days for abuse detection | Modified abuse monitoring only for customers managed by a Microsoft account team or an eligible program |
| Identity | AWS IAM; short-term Bedrock API keys | Microsoft Entra ID with role-based access; key auth can be disabled |
| Budgets | AWS Budgets alerts, updated up to three times a day; budget actions can apply an IAM policy | Budget alerts; consumption isn't stopped |
| Contract | AWS GDPR terms; Standard Contractual Clauses in the AWS Service Terms | Microsoft Products and Services Data Protection Addendum |
Where they differ most
Models
Bedrock's model list now includes OpenAI's GPT-6 and GPT-5.x next to Claude and open models. In Microsoft Foundry, Claude is a partner model where Anthropic is the seller and operator, offered hosted on Azure or on Anthropic's infrastructure.
Capacity
Both ration on-demand capacity in tokens per minute. AWS says increases aren't granted automatically; Microsoft says increases are subject to review and approval and is moving quota to subscription-level tracking across regions.
Europe
Azure's Data Zone EU follows the EU Data Boundary and can include EFTA countries. Bedrock's EU cross-Region profiles stay within the EU. On both, global options can process outside Europe.
Spend control
Neither cloud stops spend by default. Azure budgets don't stop consumption. AWS budget actions can apply an IAM policy once a threshold is passed.
Choose AWS Bedrock or Azure OpenAI
Choose AWS Bedrock if
- Your workloads, identity and procurement already run on AWS.
- Claude in Europe matters, along with GPT and open models on one bill.
- You want IAM policies to decide which roles can call which models.
Choose Azure OpenAI if
- Your company runs on Microsoft and wants Entra ID and Azure RBAC in front of every model.
- OpenAI's models are your main need.
- All AI spend must sit on your Microsoft agreement.
A third option for companies
Both lock you into one cloud's quotas and regions. Standard Compute serves Claude across several clouds, so one provider's quota or outage doesn't stop your work, and gives you GPT, Gemini and open models on the same key with Microsoft Entra sign-in.
Cost per step
Each step goes to the cheapest model that can do it and the cheapest provider fast enough to serve it.
Models
Claude, GPT and Gemini alongside open models such as GLM, DeepSeek, MiniMax and Kimi, through one API.
Processing in Europe
A Europe rule per company, team or key: every request, including retries and fallbacks, is served only by approved European inference endpoints, or refused before your prompt is sent.
Zero data retention
Required per company, team or key, and Standard Compute never trains on your prompts or outputs.
Single sign-on
Microsoft Entra ID or any SAML provider, and Claude Desktop and Claude Code sign-in with a Microsoft work account.
Budgets
A fixed monthly budget per person and app that never overruns, and spending caps per key and for the whole company.
Allow-lists
Approved providers per company, team or key, and approved models per key. Fallbacks never route around them.
Audit log
Admin changes with before and after values, and a 12-month export of every request's model, provider and region rule.
Sign our Data Processing Agreement in the dashboard, and see every sub-processor with its purpose and location.
Create a company workspaceor write to contact@standardcompute.com
Questions
Is Claude available on Azure?
Yes, in Microsoft Foundry, where Anthropic is the seller and operator and acts as an independent data processor. In Europe, Claude Global Standard is listed in Sweden Central and Data Zone Standard is not available.
Can I use OpenAI GPT models on AWS Bedrock?
AWS's Bedrock model list includes OpenAI's GPT-6, GPT-5.x and gpt-oss models.
Which keeps data in the EU?
Azure's Data Zone EU deployments follow the EU Data Boundary, and Bedrock's EU geographic cross-Region profiles keep inference within the EU. Global deployments on both may process data elsewhere.
Do Azure or AWS budgets stop spending?
Azure budgets send notifications and don't stop consumption. AWS Budgets alert, and budget actions can apply an IAM policy to stop further use.
Is there an option across both clouds?
Standard Compute is one API for Claude, GPT, Gemini and open models. Claude is served across several clouds, and each step goes to the cheapest model that can do it.
Sources (checked 9 November 2026)
- Amazon Bedrock model cardsdocs.aws.amazon.com
- Bedrock runtime quotasdocs.aws.amazon.com
- Bedrock quotasdocs.aws.amazon.com
- Bedrock Provisioned Throughputdocs.aws.amazon.com
- Bedrock cross-Region inferencedocs.aws.amazon.com
- Bedrock data retentiondocs.aws.amazon.com
- Managing costs with AWS Budgetsdocs.aws.amazon.com
- AWS budget actionsdocs.aws.amazon.com
- Bedrock API calls in CloudTraildocs.aws.amazon.com
- AWS GDPR Centeraws.amazon.com
- Azure OpenAI quotalearn.microsoft.com
- Azure OpenAI quotas and limitslearn.microsoft.com
- Foundry deployment typeslearn.microsoft.com
- Claude in Foundry: data and privacylearn.microsoft.com
- Foundry models from partners: regionslearn.microsoft.com
- Azure Provisioned Throughputlearn.microsoft.com
- Azure OpenAI limited accesslearn.microsoft.com
- Foundry Entra ID authenticationlearn.microsoft.com
- Azure budgetslearn.microsoft.com
- Microsoft Products and Services DPAwww.microsoft.com