Smarter Charts Inline Svgs And Live Usage Accounting
Source
openrouter.ai
Author
OpenRouter
Date
Context
In April 2025, OpenRouter added a way for applications to receive usage accounting alongside a streamed model response. Its announcement describes opting in through the request’s usage settings, then receiving input and output tokenThe chunk of text a model reads and writes in — roughly three-quarters of a word — and the unit AI usage is billed in.Full definition → counts, reasoning and cache details, and cost without making another API call. This puts the answer and its accounting in the same response flow, making it easier for an application to associate each completed request with the resources it consumed.
Terms in this piece · Glossary
token — The chunk of text a model reads and writes in — roughly three-quarters of a word — and the unit AI usage is billed in.
