Skip to content

TextCortex API guides

Evaluate EU data residency for an LLM API

Evaluate European model processing for your API workload. Separate model routes, application storage, logs and connected services.

Evaluate EU data residency for an LLM API

The practical takeaway

Build on a multi-model API.

For EU-hosted inference, verify the selected deployment and the systems around it. TextCortex combines EU model routes with a broader API catalog, so the route matters.

Map the request path

Start with the application receiving the user input. Identify document retrieval, logging, model inference, tool execution and result storage along the path.

A regional model endpoint addresses the model-processing step. It does not automatically determine where your own logs, database or connected services run.

Identify the model deployment

Record the actual model identifier and intended EU-hosted route for the workload. Keep processing recipients and contractual requirements with that record.

The model catalog may also include proprietary models or routes with different hosting arrangements. Do not use catalog access as proof of a single universal region.

Keep changes reviewable

Treat routing changes as deployment changes when regional requirements matter.

  • Maintain an approved set of model routes.
  • Review additions and replacements before enabling them.
  • Apply the same requirement to retries and fallbacks.
  • Retain the configuration associated with each production release.

Build with the approved route

Use the EU model access pages and TextCortex privacy policy as inputs to your review. Then validate the integration using approved test data.

This is an engineering review workflow, not a legal certification. Use the agreements and processing arrangements that apply to your organization and deployment.

Questions

Your next questions, answered.

Are all models in the catalog EU-hosted?

TextCortex offers EU-hosted model routes and a broader multi-provider catalog. Select the EU-hosted deployment required by your application; catalog membership alone does not establish where a particular request is processed.

How does model routing work?

Your application selects an available model in each API request, and TextCortex routes the request through its API. You can choose different models for different tasks. Automatic selection and fallback behavior should only be used where documented for your configuration.

Can I use an OpenAI-compatible client?

Yes. Configure the client with the TextCortex base URL and API key. Use a TextCortex model identifier and the parameters supported by that model. Model discovery, chat completions and responses are documented in the API reference.

TextCortex AI

One integration. Your choice of models.

Sign up for TextCortex, create an API key and start building with the models your application needs.

Sources and documentation

Product documentation and provider references used for this guide. Reviewed on 5 October 2026.