Skip to content

Topic

Inference token billing

How providers of hosted AI models count, price and disclose the tokens an API call consumes, including reasoning output that is charged but not returned to the caller.

Current clusters