Pricing

Prices

These rates use price card H (November 11, 2026).

See how call charges combine.

Text input is billed per token. See Inputs for formats and availability.

Worked calls

Add the lines to get a call total. These totals use price card H (November 11, 2026).

Counting and minimums

Only answered questions have a question fee. Refused questions are not charged. An all-refused request has no text, question, file or call charge.

Charged once per answered call; never on refusals, zero-answer, free, promotional or replay calls.

Several files each pay their own fee. The same text token is never charged twice.

Round any partial second up to a full second before applying the minimum billed duration.

PDF fees count physical pages. Text tokens and answered questions are charged separately. The minimum is a floor on the subtotal, not an extra fee.

Text caching

A cached text input token is a text token whose earlier processing the decision model reuses. Text caching is automatic.

An optional prompt_cache_key improves reuse within a workspace. A repeated request is not guaranteed to read from the cache.

Cache lifetime, minimum size and eviction are not published.

Caching saves USD 0.171 per million cached text tokens compared with the uncached rate.

See Text caching for usage fields, the cache hit rate and data coverage.

Credits and top-ups

Starting credit is subject to eligibility checks. Once per eligible canonical verified email address, on your first workspace, after eligibility and verification checks. Credit may be withheld or placed under review.

One credit costs USD 1. The minimum top-up is USD 5.

As a workspace owner, open Add funds to see the payment options available to your workspace.

Read Billing and credits for account terms, or follow the Quickstart to make a call.