Product
Solutions
Resources
Your LLM API and the EU AI Act
The AI Act becomes fully applicable on 2 August 2026. Run your LLM traffic on infrastructure you can explain: EU processing, per-request logging and a DPA.
On 2 August 2026 the EU AI Act becomes fully applicable. From that date, the obligations no longer sit only with the big model builders, but also with organisations that put AI systems to work in their own products and processes. For teams consuming LLMs through an API, this is the moment to answer two questions: what do we need to arrange ourselves, and what should our infrastructure arrange for us?
This page is about the second question. HostYourAI provides an OpenAI-compatible LLM API that runs entirely in the EU, with per-request logging and a data processing agreement. That does not make you compliant by itself, but it turns a series of obligations into something you can actually execute.
What the AI Act asks of teams using LLMs
Risk classes in short
The AI Act classifies AI systems by risk. A small set of applications is banned outright, high-risk systems (think CV screening or credit scoring) face heavy requirements, and most chat and text applications mainly carry transparency obligations. Which class applies depends on what you do with the model, not on which model you pick.
Transparency and documentation
Users need to know they are interacting with AI, and you need to be able to explain which model you use, with what data and what limitations. That means documenting your chain: which model, which version, where it runs and who processes the data.
Logging and traceability
For high-risk use cases logging is a hard requirement, but even outside that category you want to be able to reconstruct, after an incident, which request was processed when and by which model. Without request-level logging, every investigation starts from zero.
What your hosting choice solves, and what it does not
Your hosting provider does not decide whether your application is high-risk, and it will not write your documentation. What a good provider does do is make the infrastructure layer verifiable. If you know where your model runs, who the subprocessors are and what is recorded per request, most of your technical file can simply be filled in. If your provider is vague about those things, that gap moves into your own accountability.
EU Inference Router
‹ Insight 1 of 28 ›How HostYourAI helps
EU processing without US subprocessors
All inference runs on GPUs in European data centers, delivered by a Dutch company. There is no US cloud in the inference chain, which keeps your GDPR story simple alongside your AI Act file. Read more on our AI data sovereignty page.
An activity log for every request
Every request shows up in your activity log: timestamp, model, token counts and latency. That lets you demonstrate which model processed which request and when, exactly the kind of traceability a technical file asks for.
DPA and a public subprocessor list
A data processing agreement is ready to sign and our subprocessors are listed on a public page. Data is encrypted with AES-256 and the platform runs under a 99.9% SLA. We never train on your data.
from openai import OpenAI client = OpenAI( base_url="https://hostyourai.com/api/v1", api_key="hyai-...") client.chat.completions.create( model="llama-3.3-70b", messages=[{"role":"user","content":"Hallo!"}])
Just an OpenAI-compatible API
Nothing changes in how you build. The API speaks OpenAI's /v1/chat/completions format, and if you work with the Anthropic SDK there is a /v1/messages endpoint that accepts x-api-key authentication.
curl https://hostyourai.com/api/v1/chat/completions \
-H "Authorization: Bearer hyai-..." \
-H "Content-Type: application/json" \
-d '{
"model": "llama-3.3-70b",
"messages": [{"role": "user", "content": "Summarise this policy document."}]
}'Every call is immediately visible in your activity log.
Not legal advice
To be clear: no vendor can "handle" AI Act compliance for you, and an official "AI Act certification" for API providers does not exist today. Classification, documentation and human oversight remain the responsibility of your organisation. This page is not legal advice; involve a specialist for your specific situation. For a broader overview of European rules, see our page on AI compliance in Europe.
Questions about the AI Act and LLM APIs
Is HostYourAI AI Act certified?
No, and be wary of anyone claiming they are: no such certification exists today. What we provide are verifiable building blocks: EU processing, per-request logging, a DPA and a public subprocessor list.
Am I a provider or a deployer under the AI Act?
It depends on your role. If you build an AI system and put it on the market, you are likely a provider; if you deploy someone else's system, you are usually a deployer. Obligations differ per role, so have this assessed properly.
Does EU hosting automatically make me AI Act compliant?
No. EU hosting makes your chain explainable and keeps your GDPR position clean, but classification, user transparency and human oversight remain your work.
What is logged per request?
Timestamp, model, token counts and latency. Prompt and output content is not stored for training; see our page on zero data retention.
Does the AI Act apply if I use open-source models?
Yes. Obligations attach to the application and your role in it, not to the licence of the model.
Everything you need for AI
From model hosting to a customer-facing API, it is built for developers and businesses who want their AI running on infrastructure they actually control, inside the EU.
Your data and your models stay on European GPUs. GDPR-friendly by design.
Llama, Qwen, DeepSeek, Mistral, FLUX and plenty more. Pick one and it is warm in minutes, with no DevOps on your end.
Point your existing client at the Router and keep your tools. No rewrite, no lock-in.
From zero to a warm endpoint in minutes
No infra to manage. Pick a model, get an OpenAI-compatible URL, ship.
Pick a model
Choose from the Model Garden or paste any HuggingFace ID. Set the VRAM and pick an EU GPU.
Get your endpoint
We deploy vLLM, run readiness probes, and hand you a warm OpenAI- and Anthropic-compatible URL plus an API key.
Route and ship
Point your client at the Router. It auto-routes to a warm instance, idles GPUs when nobody is online, and logs every request.
Private by Default
HostYourAI keeps your models, prompts and data on European GPUs. It is built for teams that care about compliance, reliability and real control.
GPUs and data residency inside Europe. Your prompts never leave the EU.
Run open-weight models with no black boxes or hidden telemetry.
GPUs idle when nobody is online, so you only pay for what you run.
Your infra, your keys, your models. Leave whenever you want.
Works with the tools you already use
The Router speaks the OpenAI and Anthropic APIs, so it drops straight into the clients and SDKs your team already runs. Just change the base URL.
Try HostYourAI for free
Frequently asked questions
Can I run this in the EU?
Yes. HostYourAI runs open models on GPUs in European datacenters via vLLM. Your prompts and outputs never leave the EU and there is no US cloud provider in the chain.
Is it GDPR-compliant?
Yes. All processing happens inside the EU, a Data Processing Agreement (DPA) is available and the subprocessor list is public. Open weights also mean no training on your data.
Is the API OpenAI-compatible?
Yes. Point your existing OpenAI or Anthropic client at our Router (https://hostyourai.com/api/v1), change only the base URL and API key. No rewrite, no lock-in.
What does it cost?
Pay-as-you-go on one prepaid credit balance: the shared router per token or a dedicated GPU per hour. Free to start, no minimum, no fixed monthly fee.
Works with 390+ open models
Text and image models on dedicated EU GPUs. Every model tested on our own hardware.
Related pages
Explore more about EU-hosted AI on HostYourAI.
Lambda is winding down its Inference API. Move to HostYourAI: an OpenAI compatible API with open models on European GPUs, DPA included.
Read more →Dedicated GPU hosting without shared hardware. Full GPU capacity for your AI workloads, guaranteed performance.
Read more →Host DeepSeek R1 7B on dedicated NVIDIA A10 in European data centers. GDPR compliant, pay-as-you-go, OpenAI-compatible API.
Read more →Host Command R+ on dedicated NVIDIA A100 80GB in European data centers. GDPR compliant, pay-as-you-go, OpenAI-compatible API.
Read more →Rent NVIDIA H100 and A100 GPU servers in a European GPU cloud. Pay per hour, cancel anytime, no US cloud in the chain. Deploy a model in minutes.
Read more →Self-hosted LLM platform with managed infrastructure. The control of self-hosting without the complexity.
Read more →