Pricing rules

go2llm runs on a pay-as-you-go model with a USD balance. Current prices and the snapshot version are returned by GET /api/pricing and shown on the Pricing and billing page.

• input — input tokens and cache writes;
• output — answer tokens;
• cache read — a confirmed read of previously stored tokens;
• image models — the published price for the result or the variant.

For text, the unit is 1M tokens. Cache writes are billed at the input rate. When reliable usage is unavailable, the gateway falls back to a local estimate and does not pass it off as exact upstream accounting.

Static lists of future models and prices are not a contract: go by the live catalog and the public price list only.

Contacts: @go2llm_support and support@go2llm.tech.