About IronStratum

IronStratum is an inference platform: we run and relay AI models behind one OpenAI-compatible API at api.ironstratum.com. Point any OpenAI SDK at it, create a key in the console, and call — no adapter layer, no dialect to learn.

What we run

The catalog is small on purpose — every model in it is chosen for margin depth or uniqueness, not for the length of the list. The living catalog — aliases, prices, context windows — is published machine-readably at GET /v1/models (no key required) and rendered as the pricing table. The API itself is the price list; nothing on this page can go stale ahead of it.

Pricing doctrine

Prices are published, prepaid, and never silently changed. A model alias never changes the model that serves it; a retired model gets 90 days' notice and a named replacement. The /v1 API is additive-only — fields and endpoints may appear, never disappear or change meaning. Cached input tokens bill at a lower published rate and are reported separately in every response's usage field.

No training on your data

We never train, fine-tune, evaluate, or otherwise improve models on your prompts or completions — not by default, not with permission, not "anonymized". This is a contractual commitment in our Terms of Service, not a policy preference; what we store and for how long is in the Privacy Policy.

Reach us

Direct — create an account and the quickstart has you calling the API in minutes — or through a marketplace listing if that is where you found us. Either way the API, the prices, and the commitments on this page are the same.