Does Standard Compute support Amazon Bedrock and Google Vertex AI?

Yes. Standard Compute serves Claude from Amazon Bedrock and Google Vertex AI as well as from Anthropic and Microsoft Azure, and you can limit a key to the clouds you approve.

Can we keep Claude on Bedrock and Vertex AI only?

Yes. A provider allow-list on the company, a team or a key limits serving to the providers you choose, and fallbacks stay inside it.

Can we see which cloud served a request?

Yes. The request audit export shows the model, the provider that served it, the region rule in force and the status for every request, kept for 12 months.

What happens when one cloud is down?

Requests fail over to another approved provider of the same model automatically, inside your region and provider rules.

Can Claude stay in Europe?

Yes. Claude Opus, Sonnet and Haiku 5.5 are served from Google Vertex AI in Europe and Amazon Bedrock in Ireland. Set a Europe rule for your company, a team or a single key, and every request, including retries and fallbacks, is served only by approved European inference endpoints. If no European endpoint can serve it, the request is refused before your prompt is sent.

An AI gateway that supports Bedrock and Vertex AI in addition to OpenAI and Anthropic

Standard Compute is one API for Claude, GPT, Gemini, Mistral, Grok, DeepSeek, Qwen, GLM, MiniMax, Kimi and MiMo. It serves Claude across Anthropic, Amazon Bedrock, Google Vertex AI and Microsoft Azure, so one cloud's quota or outage does not stop your work.

You call one endpoint with one key. Standard Compute picks the model and the provider for each step, fails over between the clouds you approve, and records which provider served every request.

By Standard Compute · Last updated 9 November 2026

Which cloud serves which model

Where Standard Compute serves the main model families.

Which cloud serves which model
Models	Served by Standard Compute from
Claude Opus 5.5	Served by Standard Compute from: Anthropic, Google Vertex AI, Amazon Bedrock and Microsoft Azure
Claude Opus, Sonnet and Haiku 5.5 with a Europe rule	Served by Standard Compute from: Google Vertex AI in Europe and Amazon Bedrock in Ireland
GPT with a Europe rule	Served by Standard Compute from: Microsoft Azure in Europe
Gemini	Served by Standard Compute from: Google
Mistral Large, Grok and open models (DeepSeek, Qwen, GLM, MiniMax, Kimi, MiMo)	Served by Standard Compute from: Approved providers for each model. GLM, DeepSeek and MiniMax are also served by European providers

Source: the live model catalog at api.stdcmpt.com/v1/models and our sub-processor list, 9 November 2026.

Create a company workspaceSee pricing

A unified API for OpenAI, Anthropic and Gemini

A single API gives you frontier and open-source models, and it routes each step to the cheapest model that can do it and the cheapest provider fast enough to serve it.

The API is OpenAI-compatible, so existing SDKs and tools work by changing the base URL and key. Claude Code and Claude Desktop connect to the same account.

OPENAI_BASE_URL=https://api.stdcmpt.com/v1
OPENAI_API_KEY=YOUR_STANDARD_COMPUTE_KEY

# model ids from the live catalog
anthropic/claude-opus-5.5
openai/gpt-6.1-sol
google/gemini-4-argon
mistralai/mistral-large-4-0

Any OpenAI-compatible client

Load balance requests across GPT, Claude and Mistral

Standard Compute spreads work in two ways. Across models, routing sends each step of a task to the cheapest model that can do it, so simple steps go to fast, inexpensive models and hard steps go to a frontier model such as Claude or GPT. Across clouds, the same model is served by several providers: Claude Opus 5.5 runs on Anthropic, Google Vertex AI, Amazon Bedrock and Microsoft Azure.

If a provider goes down, your request fails over to another approved provider of the same model automatically, without leaving your region or provider rules.

Route between OpenAI and Anthropic and keep audit logs

Every admin change is recorded with before and after values, and you can export a record of every request showing the model, the provider that served it and the region rule in force, kept for 12 months.

Choose exactly which inference providers may handle your data, for the whole company, a team or a single key. Lower levels can only narrow the rules, new providers are never added without an admin's approval, and fallbacks can never route around your list.

Route between OpenAI and Anthropic and keep audit logs
Control	Scope
Provider allow-list (for example Bedrock and Vertex AI only)	Scope: Company, team or key
Model allow-list	Scope: Per key
Europe rule	Scope: Company, team or key
Request audit export: model, serving provider, region rule, status	Scope: Kept for 12 months

Questions from IT, security and finance

Does Standard Compute support Amazon Bedrock and Google Vertex AI?: Yes. Standard Compute serves Claude from Amazon Bedrock and Google Vertex AI as well as from Anthropic and Microsoft Azure, and you can limit a key to the clouds you approve.
Can we keep Claude on Bedrock and Vertex AI only?: Yes. A provider allow-list on the company, a team or a key limits serving to the providers you choose, and fallbacks stay inside it.
Can we see which cloud served a request?: Yes. The request audit export shows the model, the provider that served it, the region rule in force and the status for every request, kept for 12 months.
What happens when one cloud is down?: Requests fail over to another approved provider of the same model automatically, inside your region and provider rules.
Can Claude stay in Europe?: Yes. Claude Opus, Sonnet and Haiku 5.5 are served from Google Vertex AI in Europe and Amazon Bedrock in Ireland. Set a Europe rule for your company, a team or a single key, and every request, including retries and fallbacks, is served only by approved European inference endpoints. If no European endpoint can serve it, the request is refused before your prompt is sent.