About us

Making multi-model AI predictable

Open Vertex Router is the control layer for teams running on more than one model provider — one endpoint, routing you configure, and spend that can't run away.

What drives us

The missing control layer

Building with one model provider is a weekend project. Building with five is an operations problem.

Keys to rotate. Quotas to watch. Fallbacks to write and test. Several consoles to check, and just as many invoices to reconcile at the end of the month — usually after a runaway job has already spent the quarter's budget.

The AI tooling ecosystem caught up fast on evaluation and observability. It did not catch up on control. Teams still choose between a single provider they've outgrown and a homegrown router nobody owns.

We kept building that router ourselves, so we extracted it and now run it as a service: one endpoint, routing rules you own, budgets enforced before the request leaves, and a single invoice.

The Open Vertex way

Six positions we don't trade away

We sit in the hot path of every request our customers' products make. That position sets how we build, price and support.

Reliability

Boring on purpose

Unremarkable is the highest compliment infrastructure can get. We add milliseconds, not features nobody asked for.

Control

Hard limits, not alerts

A cap that emails you after the fact isn't a cap. Requests are pre-authorized against your budget, or they don't leave.

Pricing

A markup you can verify

Provider rates are published, and our platform fee sits on the invoice — never hidden in the token price. You can check every line.

Portability

Easy to leave

Adopting us is one line, and so is removing us. We'd rather keep the position by earning it every month.

Support

Builders answer

Questions reach the people who wrote the code. No tier-one queue between a production problem and a real answer.

Pace

Small, on purpose

We onboard fewer teams properly rather than many badly. Growth follows reliability, not the other way round.

Come build the unglamorous layer

We're a small team working on routing, budget enforcement and billing at the point where every request passes through. If that sounds good rather than tedious, we should talk.

Get in touch

Get access

Ready to route smarter?

Tell us about your stack and volumes — we'll set up your gateway, budgets and single invoice, and you'll be routing within days.