Regional Pricing

Compare total cost vs GPT-4o ($12.50/M pair). Domestic models anchor 60–90% cheaper with Smart Route, compliance & AI tools included. GPT resale at wholesale +5%. Member discounts apply to domestic models only.

Member discount applies to Chinese/domestic models only. GPT & Claude billed at list +5%.

Architecture Guidelines & FAQ

How does the Unified Token Gateway solve LLM provider fragmentation? — One OpenAI-compatible API routes to the lowest-cost or lowest-latency upstream cluster across Qwen, DeepSeek, Hunyuan and more.
What is the base URL for integration? — Use any OpenAI SDK with base_url pointing to our gateway and your dashboard secret key. Billing is per 1M tokens at the published list price.
Do caching and geo routing reduce costs? — Semantic cache lines in US, EU and Asia clusters can serve repeated prompts at up to 90% discount with live latency telemetry.
Can admins override automated pricing pipelines? — Yes — the Admin console supports manual price caps and per-model overrides for enterprise quotas.