Does ZeroEntropy support EU-only processing, and how do I select the EU instance when creating the project?

ZeroEntropy supports EU-only processing through a dedicated EU-region endpoint, so teams with strict data residency or GDPR requirements can keep retrieval and processing inside European infrastructure end-to-end. You simply target the EU API host when calling the SDK or HTTP API and configure your project to use that endpoint from day one.

Quick Answer: Yes — ZeroEntropy offers an EU-region endpoint (eu-api.zeroentropy.dev) for EU-only processing. To use it, create your project as usual, then point your SDK or HTTP calls to the EU host so all ingestion, embeddings, and search traffic stays within EU infrastructure.

Frequently Asked Questions

Does ZeroEntropy support EU-only processing for my data?

Short Answer: Yes. ZeroEntropy supports EU-only processing via a fully managed EU-based instance exposed at eu-api.zeroentropy.dev.

Expanded Explanation:
If you’re operating under GDPR or internal data residency policies, you can keep all retrieval operations within EU infrastructure by using our EU-region endpoint. When you send data (documents, chunks, queries) to eu-api.zeroentropy.dev, ingestion, indexing, embeddings, reranking, and search all run inside the EU.

The EU endpoint is designed for teams in regulated industries (finance, healthcare, public sector, legal) that can’t ship data outside the EU but still need high-recall hybrid retrieval, calibrated rerankers, and predictable latency for production RAG and agent workloads. You get the same retrieval stack—zembed-1 embeddings, zerank-2 rerankers, and the unified Search API—just hosted in EU infrastructure.

Key Takeaways:

• EU-only processing is supported via the eu-api.zeroentropy.dev endpoint.
• All ingestion and query traffic sent to this host is handled in EU-region infrastructure.

---

How do I select the EU instance when creating a ZeroEntropy project?

Short Answer: Use the same dashboard and API key flow, then configure your application to call the EU endpoint (https://eu-api.zeroentropy.dev) instead of the default global host.

Expanded Explanation:
There’s no separate “EU project type” you have to create. Instead, EU residency is enforced at the network level: your app uses the same API key, but all traffic is directed to the EU host. From your perspective, this is an API swap—change the base URL and your project becomes EU-only in practice.

This keeps onboarding simple. You sign up at dashboard.zeroentropy.dev, grab your key, and then decide whether your integration should talk to the global or EU instance. For GDPR-sensitive deployments, you hard-code or centrally configure the EU endpoint in your environment variables and SDK settings and keep it consistent across ingestion and query paths.

Steps:

1. Create an account & project at dashboard.zeroentropy.dev and generate an API key.
2. Configure your base URL to https://eu-api.zeroentropy.dev in your SDK or HTTP client (instead of the default API host).
3. Ingest and query via the EU host so embeddings, hybrid retrieval, and reranking all run within EU infrastructure.

---

What’s the difference between the EU instance and the default ZeroEntropy API?

Short Answer: Functionally they’re the same retrieval stack; the main difference is where your data is processed and stored (EU infrastructure vs. global/default).

Expanded Explanation:
The EU instance exposes the same core capabilities—zembed-1 embeddings, zerank-2 rerankers, and the Search API with dense + sparse + rerank in a single endpoint. The distinction is infrastructure locality and compliance posture. The EU endpoint is designed for teams that need to prove data stays in the EU while still getting high NDCG@10, stable p99 latency, and token-efficient RAG.

From an integration standpoint, the only change is the base URL. From a governance standpoint, your DPO or compliance team can rely on the fact that all traffic to eu-api.zeroentropy.dev is handled inside EU-region infrastructure rather than transiting outside the region.

Comparison Snapshot:

• Default API: Global infrastructure, ideal for teams without strict data residency constraints and latency-sensitive workloads distributed worldwide.
• EU Instance (eu-api.zeroentropy.dev): EU-region infrastructure, optimized for GDPR-sensitive or EU-only data policies.
• Best for: Teams that must keep all retrieval processing inside the EU but still want a unified hybrid + rerank stack.

---

How do I configure my SDK or HTTP client to use the EU endpoint?

Short Answer: Set the base URL to https://eu-api.zeroentropy.dev in your client configuration and keep all ingestion and query calls pointed there.

Expanded Explanation:
ZeroEntropy is designed to be an API swap, not a platform migration. Once you have your API key, you choose which host to hit. For EU-only processing, treat the base URL as a mandatory configuration: environment variables, infra-as-code, and CI/CD templates should all reference eu-api.zeroentropy.dev to avoid accidental cross-region calls.

If you’re building RAG or agent pipelines, make sure both ingestion (embeddings, indexing, OCR-backed ingestion) and online retrieval (search + rerank) share the same EU base URL. That’s how you maintain end-to-end EU residency—from raw documents, through zembed-1 embeddings, to zerank-2-driven rankings.

What You Need:

• An API key from dashboard.zeroentropy.dev with access to embeddings/rerank/Search API.
• Client configuration (SDK or HTTP) that sets https://eu-api.zeroentropy.dev as the base URL for all requests.

---

How does EU-only processing fit into ZeroEntropy’s broader security and compliance posture?

Short Answer: EU-only processing complements ZeroEntropy’s SOC 2 Type II and HIPAA readiness by adding a clear data residency control for teams with regional constraints.

Expanded Explanation:
Security and compliance are first-class design constraints for ZeroEntropy’s retrieval stack, not afterthoughts. We’re SOC 2 Type II and HIPAA ready, which means our controls around access, logging, and data handling have been independently audited. For many enterprise teams, that covers the “how” of processing; EU-only processing addresses the “where.”

By routing traffic through eu-api.zeroentropy.dev, you align retrieval workloads with GDPR and internal residency policies while still benefiting from calibrated rerankers, hybrid dense/sparse retrieval, and predictable p50–p99 latency. This is especially relevant for legal, clinical, and financial stacks where shipping documents or logs outside the EU is a non-starter, but RAG and agents still need human-level retrieval performance.

Why It Matters:

• Regulatory fit: EU-only processing plus SOC 2 Type II and HIPAA readiness reduces friction with legal, security, and compliance reviews.
• Production reliability: You don’t trade off compliance for performance—EU-region deployments still deliver strong NDCG@10 and stable p99 latency for real traffic.

---

Quick Recap

ZeroEntropy does support EU-only processing via the eu-api.zeroentropy.dev endpoint. You create your account and project at dashboard.zeroentropy.dev, grab an API key, and then point all ingestion and query traffic at the EU host. Functionally, you get the same retrieval stack—zembed-1 embeddings, zerank-2 rerankers, and hybrid search in a single API—but processed entirely within EU infrastructure, aligned with SOC 2 Type II, HIPAA readiness, and GDPR-conscious data residency requirements.

Next Step

Get Started