Frontier models. 30% off the list.
Every rate below is exactly what you pay — a flat 30% under the lab's list price for input, output, cached and reasoning tokens. No markup, no rounding, no fine print.
| Model | Context | Max out | Input / M | 5m write / M | 1h write / M | Cache hit / M | Output / M |
|---|---|---|---|---|---|---|---|
| Loading pricing... | |||||||
No subscription
30% cheaper than going direct
Cancel any time