Skip to content

02 Control Keys · Budgets · Guardrails

Set the rules once.
Every request follows.

Keys, budgets and guardrails per team, and personal data stopped before any model sees it.

  • GLM
  • DeepSeek
  • Qwen
  • Kimi
  • Mistral
  • Llama
  • gpt-oss
  • Gemma
  • Loes
  • Nemotron
  • Claude Code
  • Codex
  • OpenCode
  • Cursor
  • Cline
  • Loes
  • OpenAI SDK
  • Anthropic SDK

What it does

One router per team,
with its own rules.

Every team gets its own router. You decide which models it may use, how much it may spend each month and whether its traffic stays in Europe. Every key you share with the team follows those rules, in the dashboard and through the API.

Guardrails check each request before a model is chosen. Personal data, prompt injection, moderation and your own blocked terms can each be switched off, flagged or blocked, per team. The prompt text itself is never stored.

The route decides who answers: your own GPUs first, a model you picked for the work, or the router's choice of the best open model for code, analysis or plain writing. When a budget is used up, new requests stop at the limit instead of running on.

Read more

Example of a team router

Monthly budget

€1,240of €2,000

62% used

Route
Best model for code
Personal data
flag
Europe only
off
Models
GLM 5.3 Kimi K3 Claude, on your key
API Routers
$ curl hostyourai.com/api/v1/teams/platform/router
  router  platform
  budget  €2,000 a month, stop at limit
  route  best model for code
  pii  flag before any model
  inject  block
  region  europe, plus your own keys
  ✓ 14 keys now follow this router
Why teams use it

Rules that hold
on every request.

Budgets per team and per model

A monthly budget for the team and a cap per model inside it. Spend shows per team, per key and per model in Activity.

Europe only, when you need it

Switch it on and a key only reaches machines in European data centres. The strictest rule of key and team always wins.

Personal data checked first

Names, account numbers and other personal data are checked before a model is chosen. Each team decides whether such a request is flagged or blocked.

Scriptable from end to end

Teams, routers and keys have their own API, so a new team can be set up from your own tooling with a personal key.

Screenshot of the team budgets: each team's spend this month against its monthly budget
Questions

Questions we get

Do our existing keys keep working?

Yes. A key you share with a team follows that team's router from then on. Keys outside a team keep the settings you gave them.

What happens when a team reaches its budget?

New requests from that team get a clear budget error until the next month starts or you raise the budget. Requests that are already running finish.

Can we keep our own contracts with model vendors?

Yes. A model on your own vendor key can sit in a team's route with a monthly token cap. Those tokens are billed by your vendor, under your own contract.

The platform

One platform, four layers

Start with one layer and add the others when the work asks for it. They share one key, one balance and one Activity view.

Code

Your coding agent on open models in Europe, set up with one command.

Open

Control

Keys, budgets and guardrails per team, and personal data stopped before any model sees it.

Open

Serve

Shared per token, or running for you alone, from European data centres.

Open

EU compute

Private GPUs in Europe, H100 to B300, sized for response time, generation speed and concurrent users.

Open

Set up your first
team router.

Create a key, add your team and choose its rules. Every request follows them from the first call.

  • Jantine
  • Hans
  • Sander
  • Jos
  • Daan
  • Kyra
  • Margriet
  • Stan
The team behind HostYourAI