Commit early. Pay less.
Buy inference capacity in advance at a significant discount: a fixed block of tokens on a named model for a delivery month, served through one OpenAI-compatible key. Draw it down at your pinned per-token rates from the month’s first day.
- Up to 20% off list for committing ahead
- A posted price, pinned the moment you buy
The discount is a curve, not a negotiation
The standard curve prices each delivery month below the model’s list rate (its posted per-1M output rate, fixed at listing): 10% per 30 days to maturity, the month’s last day, capped at 20%. The cap binds from about 60 days before the month ends, so a contract bought two months ahead earns the full discount. Live prices are in the market.
Prefer your own price? Rest a funded bid below the ask on the order book; it fills when a seller meets it.
Size, and models that haven’t shipped
Need more than the posted book shows? Post the size, months, and models you need on the RFQ (request-for-quote) desk; providers respond with quotes, and awarding one is binding on the spot. RFQ deals settle directly between you and the provider. Benchmark-referenced contracts cover models that haven’t shipped: quoted in Synthetic Tokens and resolved to a named model shortly before delivery. The mechanics: the FAQ.