About Us — MCLY Technology

Last updated September 2026

What we do

MCLY Technology runs a control plane for large language models. Instead of signing separate agreements with every frontier lab, our customers hold one contract, receive one API key and send every request to a single OpenAI-compatible endpoint. We handle authentication, provider selection, retries, failover and metering behind that endpoint.

For teams that run their own weights, we also reserve dedicated GPU capacity — RTX 4090, A100 and H100 instances, from a single card to eight-GPU clusters — with root access and persistent volumes, billed monthly.

How we work

We are an engineering-led company. The people who answer your integration questions are the same people who operate the routing layer, which is why our support replies are specific rather than scripted. Requests are metered at the token level against the upstream provider's own accounting, so what you see in your dashboard is what the provider billed.

Principles

No lock-in. Our API follows the OpenAI specification. If you leave, you change a base URL and a key — nothing else about your codebase.
Transparent metering. Per-request token detail, per-key burn-down and per-project attribution are available on every plan, including the smallest.
Boring reliability. Failover, retry and health checks are defaults, not premium features you have to configure.
Clear pricing. A published rate card, invoices in USD, and no charge for requests that fail inside our gateway.

Registered office

MCLY TECHNOLOGY INC
4770 Baseline Rd Suite 228, Boulder, CO 80303, USA
[email protected]

Talk to us

Evaluating the gateway for a production workload, or sizing a GPU reservation? Send us your requirements and we will come back with a concrete recommendation — including the option to run a paid pilot against your own traffic.