Gemini with EU data residency
Route Gemini models through Google Vertex AI EU regions. The full request path stays in Europe.
https://router.eu.requesty.ai/v1The consumer Gemini API processes data globally. For guaranteed EU processing, Google offers Gemini through Vertex AI with region pinning — including EU regions such as Belgium (europe-west1), Netherlands (europe-west4), Finland (europe-north1) and Warsaw (europe-central2).
Requesty exposes those Vertex EU deployments behind one OpenAI-compatible endpoint, routed through our Frankfurt gateway with zero data retention — no Google Cloud project, service-account JSON or per-region setup required.
Gemini models available in EU regions
Live from the Requesty model catalog. Prices per 1M tokens. Regions are merged per model.
| Model | Platform | EU regions | Input /1M | Output /1M | Context |
|---|---|---|---|---|---|
| Gemini 3.5 Flash | Google Vertex AI | EU multi-region | $1.50 | $9.00 | 1.0M |
| Gemini 2.5 Flash Image | Google Vertex AI | Belgium, Finland, Madrid, Milan, Netherlands, Warsaw | $0.30 | $2.50 | 1.0M |
| Gemini 3.1 Flash Lite | Google Vertex AI | EU multi-region | $0.25 | $1.50 | 1.0M |
Pricing matches provider list prices. Full catalog on the models page.
One base URL. Full EU residency.
OpenAI SDK compatible — point it at the EU gateway and pick an EU-region model.
Frequently asked questions
Yes, through Google Vertex AI with region pinning. The consumer Gemini API processes data globally, but Vertex AI lets you pin Gemini processing to EU regions such as europe-west1 (Belgium) and europe-west4 (Netherlands). Requesty routes to these EU deployments through one OpenAI-compatible API.
If the goal is EU data residency for Gemini and other frontier models without managing Google Cloud, an EU-hosted gateway is the practical alternative. Requesty runs in Frankfurt, routes to Vertex AI EU regions (plus Bedrock and Azure EU regions for Claude and GPT), and replaces per-provider setup with one API.
Gemini 3.5 Flash, Gemini 3.1 Flash Lite and Gemini 2.5 Flash Image have EU-region Vertex deployments routable through Requesty, pinned to regions including Belgium, Netherlands, Finland, Warsaw, Madrid and Milan.
Not with Requesty. We operate the Vertex AI EU deployments; you call one OpenAI-compatible API with your Requesty key and request a model such as vertex/gemini-3.5-flash@eu. No GCP project, IAM setup or service-account keys.
Yes. The Requesty gateway processes requests in Frankfurt with zero data retention and a DPA is available. Combined with Vertex AI EU-region processing, prompts stay under EU jurisdiction for the entire request path.
EU data residency for other model families: Claude · OpenAI GPT · Mistral — or see the full EU AI Gateway and our Data Processing Agreement.
Run Gemini in the EU today
One base-URL change. $10 free credits, no credit card required.