About Rhodes
Rhodes is a prepaid inference API that gives you a hard spend cap. We built it because every other inference API sends you a bill at the end of the month, and there is no way to know what that bill will be until it arrives.
With Rhodes, you buy credits before you spend them. When they run out, the gateway stops serving and tells you so. That is the hard cap, and the amount you topped up is the whole of it.
We speak the OpenAI chat-completions API, so existing clients work without modification. Open-weight models at a flat per-million-token rate and frontier models through the same key, the same endpoint, and the same balance.
Why we built this
Every developer we talked to had the same complaint: inference bills are unpredictable. A bug that retries forever, a user who sends a 50k-token document, or a spike in traffic can turn a $50 bill into a $5,000 bill. With Rhodes, you know exactly what you have left, always.