gpt-4.1-mini

Score

79

Task done

92.63%

80–100% 95% CI bootstrap over items · n=95 · measured 2026-10-04

Refusal quality

83.3%

names the gap 80% · offers a route 80% · 10 probes x 3 iterations · 3 silent turns scored zero

Fabrication

3%

1 of 30 probe runs

Dead-air

61.4%

167 of 272 turns

Tool silence

97.1%

167 of 172 tool-call turns

Stalled

2.2%

2 of 90 runs

TTFT p50

480ms

480–755 p50–p90 · n=38

Cost / 1M tok

$1.60