The same open models behind the same kind of API, running on European GPUs under European law.
Together AI has earned its place among the serious players in AI infrastructure. It calls itself the "AI Native Cloud" and covers the full stack: serverless inference on open models, fine-tuning, and entire GPU clusters for training. None of that is in dispute. The question European teams keep running into is a different one: where does my workload actually run, and whose law applies to it?
Together AI is a US company. For a side project that hardly matters, but the moment personal data or confidential business information flows through your prompts, jurisdiction becomes an architecture decision. HostYourAI exists for exactly that case: the same open models behind an OpenAI compatible API, served from GPUs in European datacenters by a Dutch company.
Credit where it is due: Together AI operates at real scale and has a strong research pedigree, with meaningful contributions back to the open source ecosystem. The platform spans everything from a single API call to a training cluster. If jurisdiction plays no role in your decision, it is a perfectly reasonable choice.
So the argument is not that Together AI does something wrong. The argument is that a European organisation with a DPO, a record of processing activities and customers asking about data residency inherits a discussion with a US vendor that simply does not exist with a European one.
A US provider is subject to US law regardless of what the contract says or where the servers stand. US authorities can compel access to data held abroad. Since Schrems II, that is precisely the point privacy lawyers test transfers against, and contractual clauses do not make it go away.
Every prompt sent to a US API is an international data transfer that GDPR requires you to assess, document and monitor. Keep everything inside the EU and that entire exercise disappears from your records of processing.
In healthcare, finance, legal work and government, "data stays in the EU" is increasingly a procurement requirement rather than a preference. You want a vendor that can demonstrate it, not approximate it.
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!"}])
HostYourAI is a Dutch company and all inference runs in European datacenters, with no US cloud anywhere in the chain. You get a DPA, a public subprocessor list, and a guarantee that your data is never used for training. The platform carries a 99.9% SLA.
The API speaks the same /v1/chat/completions format as OpenAI, including SSE streaming. Teams on the Anthropic SDK get a /v1/messages endpoint with x-api-key authentication. Existing code, LangChain pipelines and tooling keep working.
You choose per workload: pay-per-token on the shared Router with scale-to-zero, or a dedicated GPU instance running vLLM exclusively for you. The model catalog lists open models such as Llama, Qwen, Mistral, DeepSeek and GLM; switching is a one-string change. Billing is prepaid with no subscription; see the pricing page.
from openai import OpenAI
client = OpenAI(
base_url="https://hostyourai.com/api/v1",
api_key="hyai-...",
)
response = client.chat.completions.create(
model="llama-3.3-70b",
messages=[{"role": "user", "content": "Summarise this customer call."}],
)
The longer walkthrough, including how to pick an equivalent open model, lives in the guide Migrate your OpenAI client to the EU Router.
Yes. You change the base URL, the API key and the model name. Streaming, temperature and system prompts behave the way you expect.
Largely, yes. The catalog covers the major open families such as Llama, Qwen, Mistral and DeepSeek. Anything missing can be deployed yourself on a dedicated instance with vLLM.
No, it is not a training platform. If you fine-tuned a model elsewhere, you can serve those weights on a dedicated instance inside the EU.
Prepaid credit, billed per token on the Router or per hour on a dedicated instance. No subscription and no minimum commitment.
Yes. There is a 99.9% SLA, a per-request activity log, and optional RAG injection from a linked knowledge base.
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.
No infra to manage. Pick a model, get an OpenAI-compatible URL, ship.
Choose from the Model Garden or paste any HuggingFace ID. Set the VRAM and pick an EU GPU.
We deploy vLLM, run readiness probes, and hand you a warm OpenAI- and Anthropic-compatible URL plus an API key.
Point your client at the Router. It auto-routes to a warm instance, idles GPUs when nobody is online, and logs every request.
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 freeIf a US cloud is off the table, HostYourAI gives you the same developer experience on European infrastructure.
Citizen data that legally has to stay in the EU, with full auditability.
Finance, healthcare and legal teams under GDPR, DORA and the AI Act.
Ship AI features your customers trust, without a US sub-processor.
Deliver private AI for clients on infrastructure you can stand behind.
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.
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.
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.
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.
Text and image models on dedicated EU GPUs. Every model tested on our own hardware.
Explore more about EU-hosted AI on HostYourAI.
Deploy open source LLMs with one click. Llama, Mistral, DeepSeek, Qwen and 100+ other models ready to use.
Read more →AI for financial institutions on European GPUs. Data stays in the EU, with open models, a DPA and attention to DORA and GDPR.
Read more →A ChatGPT alternative that runs on open models and European GPUs. OpenAI-compatible API, GDPR-friendly and without your data leaving the EU.
Read more →Host Phi 3 Medium on dedicated NVIDIA A10 in European data centers. GDPR compliant, pay-as-you-go, OpenAI-compatible API.
Read more →Private LLM hosting on dedicated hardware. No shared resources, full data isolation, your own AI infrastructure.
Read more →Host DeepSeek R1 70B on dedicated NVIDIA A100 80GB in European data centers. GDPR compliant, pay-as-you-go, OpenAI-compatible API.
Read more →