Total API requests served through OpenRouter — measures how many individual queries each model handles.
Indicates breadth of adoption by counting many small interactions instead of only large token-heavy sessions; responds quickly to demand changes after launches or pricing updates.
High request counts can reflect low price or default routing rather than quality preference; request-level metric ignores response quality and task difficulty.
Higher is better for request-level popularity. Some models have fewer requests but higher token volume due to long-context workloads.
High request counts can reflect low price or default routing rather than quality preference.
No scores available for this benchmark yet.