Token Usage / Pricing

What it is

Token-based pricing charges for AI API usage based on the number of tokens processed—both in your input (prompt tokens) and the output (completion tokens). Prices are typically quoted per million tokens and vary by model, with more capable models costing more.

Why it matters

Understanding token pricing helps you estimate and control costs. Input and output tokens often have different prices. Longer prompts and longer responses cost more. For high-volume applications, optimising prompt length and choosing appropriately-sized models significantly impacts costs.