AI capture spend
Tokens and cost for reading documents, per period, and anything the price list could not price.
The question it answers: what has letting Orin read documents cost us?
Capture is priced per document by how much text it contains, so this report is the bill.
Unpriced items
Some usage cannot be priced from the published list: an unfamiliar model, or a format with no rate. Those are shown separately rather than folded into the total, because a figure that was estimated must not be presented as one that was measured.
Spend and the limit that caps it both live in Capture settings. The limit is what stops the number growing; the spend is what tells you where to set it.
Common questions
The total looks high for one month. A bulk upload reads every document in it. Capture cost follows volume of paper, not number of clients.
Was this page helpful?
Still stuck?