Product
Solutions
Resources
Chatbot Hosting Platform
Build and deploy AI chatbots with full GDPR compliance.
Every chatbot is two things: a conversation your users see, and a model call your server makes. HostYourAI takes care of the second half. You keep the UI, the conversation logic and the tone of voice; the platform serves open models like Llama, Qwen, Mistral and Gemma with vLLM on European GPUs behind an OpenAI-compatible API. Your bot, your infrastructure, no black box.
Chat is where users are most candid. They paste order numbers, describe health complaints, quote contracts. A customer-facing or internal chatbot therefore processes exactly the data your privacy policy promises to protect, and where the model runs decides whether you can keep that promise.
Why chatbots on your own infrastructure in the EU
Conversations are regulated data
Chat transcripts routinely contain names, account details and personal circumstances, which makes them personal data under the GDPR. When completions run in European datacenters, with TLS in transit, AES-256 at rest and a Data Processing Agreement, your bot's data flow matches what your legal team already signed off. HostYourAI never trains on your users' conversations.
Procurement says yes faster
Public-facing bots for European organizations often stall in review because the model endpoint sits outside the EU. An endpoint on European infrastructure with a public subprocessor list and a 99.9% uptime SLA answers the questionnaire as asked, which matters doubly in sectors like finance and government.
Costs that scale with conversations
Chatbot traffic is spiky: quiet nights, busy Mondays. Pay per token on the shared Router while volume is variable, and switch to a per-hour dedicated GPU instance once load is steady enough to justify reserved capacity. Both draw from one prepaid credit balance, explained on the pricing page.
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!"}])
From model to chatbot in practice
Four pieces make a working bot. The endpoint: /v1/chat/completions with SSE streaming, so replies render word by word. The system prompt: your bot's role, tone, boundaries and escalation rules, versioned in your repo. The knowledge base: link one to the Router or your instance and answers are grounded in your own documents automatically, with source citations on non-streaming responses. The frontend: your website widget, app screen or intranet page, storing conversation history on your side. HostYourAI does not sell a chatbot product; it serves the model yours is built on, and you can prototype prompts in the Playground before writing any code.
| Chatbot component | Your side or platform side |
|---|---|
| Chat UI and conversation history | Yours |
| System prompt and escalation rules | Yours |
| Model serving and streaming | Platform |
| Document grounding with citations | Platform, via the knowledge base link |
| User authentication and rate limits | Yours |
How to start
Pick a model in the Model Garden, create a key, and wire your backend to the endpoint. This JavaScript handler turns a user message into a streamed bot reply with Gemma 3 12B:
import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://hostyourai.com/api/v1",
apiKey: "hyai-...",
});
export async function reply(history, userMessage) {
return client.chat.completions.create({
model: "gemma-3-12b",
messages: [
{ role: "system", content: "You are the support assistant for Acme. Be brief, friendly and honest. Escalate billing disputes to a human." },
...history,
{ role: "user", content: userMessage },
],
stream: true,
});
}Questions about chatbots on your own infrastructure
Is this a ready-made chatbot product?
No. HostYourAI serves open models on EU GPU infrastructure; you build the bot on top with any stack you like. The OpenAI-compatible API means most chat frameworks work by changing the base URL.
How does the bot know my company's information?
Link a knowledge base with your documents to the Router or your instance. Relevant context is injected into each completion automatically, and non-streaming responses include the source documents, as covered in Build RAG on EU GPUs.
Can replies stream like typing?
Yes. The endpoint supports SSE streaming, so your frontend can render tokens as they arrive, which is what makes a bot feel responsive.
Which model should a chatbot use?
Smaller models keep costs low for straightforward support flows; larger ones like Llama 3.3 70B handle nuance and multilingual users better. Try candidates in the Playground before committing.
Where are conversations processed and stored?
Completions run in European datacenters across multiple European regions. Conversation history lives in your own systems; the platform is not a system of record for your chats.
Can I move an existing OpenAI-based bot over?
Usually in minutes: swap the base URL and key, then pick an open model. The guide Migrate your OpenAI client to the EU Router walks through it.
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.
Point your Anthropic client at our Router through a /v1/messages shim and run open models on European GPUs. Drop-in, EU-hosted, no rewrite.
Read more →Fine-tune LLMs for your specific use case. LoRA, QLoRA, full fine-tuning on dedicated GPUs.
Read more →One OpenAI- and Anthropic-compatible endpoint that routes your requests to open models on European GPUs. Drop-in, EU-hosted, no lock-in.
Read more →Browse a curated catalog of serveable open chat, image and embedding models with live warm/EU state. Deploy in one click on European GPUs.
Read more →Deploy any open model on a dedicated European GPU with vLLM. Your own endpoint, encrypted upstream keys, idle when nobody is online.
Read more →A drop-in OpenAI-compatible API on European GPUs. Change only the base URL and run open models, GDPR-friendly, without rewriting your code.
Read more →