pub fn itl_ms(decode_window_ms: f64, output_tokens: usize) -> Option<f64>Expand description
2026-09-26: Inter-token latency: decode_window_ms / (output_tokens − 1).
The one definition both clocks use: the client passes its first-delta →
stream-end window, and ChatOutcome::server_tpot_ms passes the
endpoint’s usage.decode_time_ms.
None, never 0, below two output tokens and for a window that is not
positive and finite: a zero would read as an infinitely fast decode.