---
title: "Claude Opus 5 in the EU: API hosted in Europe"
description: "Use Anthropic's Claude Opus 5 on inference.eu: requests processed in EU data centres, one data processing agreement, and nothing used for training."
canonical: "https://inference.eu/models/claude-opus-5/"
language: "en"
---

Reasoning · Anthropic

# Claude Opus 5 in the EU, processed in European data centres

Claude Opus 5 is Anthropic's large frontier model, released in July 2026, with a one-million-token context window and adaptive thinking. On inference.eu you call it with the Anthropic Messages or OpenAI request format, your requests are processed in EU data centres, and nothing you send is used for training.

[Contact sales](https://inference.eu/contact/)[Example request](#example)

1.  01Context window: 1,000,000 tokens
2.  02EU data centres
3.  03Proprietary

## What Claude Opus 5 is

Claude Opus 5 is a closed frontier model from Anthropic, released on 24 July 2026 as the successor to Claude Opus 4.8. It reads text and images and writes text. Its context window is one million tokens, and a single response can be up to 128K tokens long. Anthropic gives its reliable knowledge cutoff as May 2026.

Opus 5 uses adaptive thinking: the model decides how much to reason before it answers, and an effort setting steers how deep it goes. Anthropic's default effort for Opus 5 is high. Anthropic does not publish the parameter count or the weights, so the model is only available as a service.

## What it does well, and where it falls short

Anthropic reports the largest gains over Opus 4.8 in agentic coding, computer use and long-horizon knowledge work. Opus 5 is much better at checking its own work and iterating until a task succeeds, which matters most in long agent runs where one wrong step can derail the rest.

It is a large model, so it is slower and costs more per token than Claude Sonnet 5.5 or Claude Haiku 5.5. For short, well-defined tasks such as classification or simple extraction, a smaller model is usually the better choice. Anthropic has since released Claude Opus 5.5 and lists Opus 5 as a legacy model that remains available.

## Calling Claude Opus 5 through the API

inference.eu speaks the Anthropic Messages format, so the official Anthropic SDKs work once you change the base URL and the key. Claude Code works the same way: set `ANTHROPIC_BASE_URL` to the inference.eu host, `ANTHROPIC_AUTH_TOKEN` to your inference.eu key, and choose `claude-opus-5` as the model.

If your code uses the OpenAI SDKs, LangChain, LlamaIndex or the Vercel AI SDK, you can call the same model through the OpenAI-compatible endpoint with the model set to `claude-opus-5`. Streaming, tool calls and image input work in both formats, so you can keep the client you already have.

## Claude Opus 5 with data processed in the EU

When you call Claude Opus 5 on inference.eu, your prompts, images and the model's responses are processed in data centres in the European Union. They are used to answer your request and are never used to train any model.

We sign one data processing agreement under Article 28 of the GDPR with every customer and provide our list of subprocessors, so your data protection officer can see who handles the data and where. That one agreement covers Claude Opus 5 and every other model on inference.eu. Billing is in euros.

## Claude Opus 5

Maker

Anthropic

Made in

United States

Type

Reasoning

Context window

1,000,000 tokens

Licence

[Proprietary](https://www.anthropic.com/legal/commercial-terms)

Released

July 2026

Model ID

claude-opus-5

Processing

EU data centres

Training on your data

Never

Source

[platform.claude.com](https://platform.claude.com/docs/en/models/opus-5/overview)

claude-opus-5 · Python

```
import os
from openai import OpenAI

client = OpenAI(base_url="https://api.inference.eu/v1", api_key=os.environ["INFERENCE_EU_API_KEY"])

response = client.chat.completions.create(
    model="claude-opus-5",
    messages=[{"role": "user", "content": "Summarise this contract in five bullet points."}],
)
print(response.choices[0].message.content)
```

## What teams use it for

-   1

    ### Long-running coding agents

    Run Claude Code or your own agent on large repositories, with Opus 5 planning changes, running tools and checking its own results over many steps.

-   2

    ### Complex knowledge work

    Draft and review long reports, contracts and analyses where the model has to keep many documents and constraints in view at once.

-   3

    ### Computer use and screen tasks

    Let an agent read screenshots of web apps and internal tools and decide what to do next, with requests processed in the EU.

## Claude Opus 5: common questions

The questions buyers, developers and legal teams ask us most. Something missing? Ask us directly.

Yes. On inference.eu, Claude Opus 5 is served from EU-only endpoints and your requests are processed in data centres in the European Union. You get one data processing agreement for all models and one invoice in euros.

Whether your use is lawful depends on what you process and why, not on the model alone. On inference.eu, requests are processed in EU data centres, are not used for training, and are covered by a data processing agreement under Article 28 of the GDPR with a published subprocessor list. That gives you the contractual basis to use Opus 5 with personal data.

No. What you send to inference.eu is used only to produce your response. It is not used to train or fine-tune Claude or any other model.

Claude Opus 5 has a context window of one million tokens and can write up to 128K tokens in one response. The system prompt, messages, images, tool definitions and the model's own thinking all count towards the window.

Yes. Opus 5 accepts text and images as input and produces text. You can send screenshots, charts, photos and scanned pages in the same request as your text. It does not generate images.

Yes. Set \`ANTHROPIC\_BASE\_URL\` to the inference.eu host and \`ANTHROPIC\_AUTH\_TOKEN\` to your inference.eu key, then select \`claude-opus-5\` as the model. Your source code and prompts are then processed in EU data centres.

Only the base URL, the API key and the model id. inference.eu speaks the Anthropic Messages format, so the Anthropic SDKs, your prompts, tools and streaming code stay as they are. If you prefer the OpenAI SDKs, the same model is available through the OpenAI-compatible endpoint too.

No. Claude Opus 5 is a closed model. Anthropic does not publish the weights or the parameter count, and use is governed by Anthropic's terms. If you need a model with open weights, look at gpt-oss-120b or DeepSeek V4.1 Flash on inference.eu.

Opus 5 is the larger model, aimed at long, difficult tasks that need careful judgement. Sonnet 5.5 is newer, faster and cheaper per token, and Anthropic positions it for well-scoped everyday coding and agent work. Many teams use Sonnet 5.5 by default and send the hardest tasks to Opus.

Yes. Anthropic has released Claude Opus 5.5 and now lists Opus 5 as a legacy model. Opus 5 remains available, and Anthropic has said it will not retire it before July 2027.

[Ask a question](https://inference.eu/contact/)

## Related

[All models](https://inference.eu/models/#directory)

-   [Claude Sonnet 5.5Use Anthropic's Claude Sonnet 5.5 on inference.eu: requests processed in EU data centres, one data processing agreement, nothing used for training.Chat](https://inference.eu/models/claude-sonnet-5-5/)
-   [Claude Haiku 5.5Use Anthropic's Claude Haiku 5.5 on inference.eu for high-volume tasks: requests processed in EU data centres, one DPA, nothing used for training.Chat](https://inference.eu/models/claude-haiku-5-5/)
-   [gpt-oss-120bRun OpenAI's open-weight gpt-oss-120b reasoning model on inference.eu: the OpenAI API format, with requests processed in EU data centres, billed in euros.Reasoning](https://inference.eu/models/gpt-oss-120b/)
-   [DeepSeek V4.1 Flash (1017)Run DeepSeek V4.1 Flash (1017) on inference.eu: a multimodal open reasoning model with 1M context, OpenAI-compatible, processed in the EU, not in China.Reasoning](https://inference.eu/models/deepseek-v4-1-flash-1017/)

## Move your AI workloads to EU infrastructure

Tell us which models you use and roughly how much traffic you have. We reply within one working day with pricing and next steps.

-   Reply within one working day
-   Pricing for your volume
-   DPA ready to sign