No markups, no hidden platform fees. Connect to the world's leading models directly at raw vendor wholesale cost, with built-in dynamic caching and smart failover routing.
Real-time pricing for token exchange rates as of June 2026.
| Model ID & Creator | Modality | Input / 1M Tokens | Output / 1M Tokens | Context Window | Status |
|---|
Choose the routing capacity that suits your operational scale.
Best for personal projects and local development. Start testing current June 2026 models without a credit card.
Designed for high-frequency production apps, with automatic failover, lower concurrent latency, and dedicated developer support.
Built for high-throughput enterprises, with dedicated routing nodes, isolated compliance policies, and custom SLAs.