
== o4-mini-2025-04-16_telecom_default: 192 successful, 264 failed runs, 27.0 s
   successful runs flagged   figured   0.5%  baseline   0.5%
   failed runs flagged       figured   3.4%  baseline   2.7%
   corruptions caught        figured 562/562  baseline 562
   substitutions caught      figured 0/445  baseline 0
   calls to unknown tools    successful runs 0  failed runs 0
   real wrong values flagged 0/36  baseline 1; unneeded writes flagged 2/330

== gpt-4.1-mini-2025-04-14_telecom_base: 200 successful, 256 failed runs, 25.7 s
   successful runs flagged   figured  41.5%  baseline  41.5%
   failed runs flagged       figured  48.4%  baseline  48.4%
   corruptions caught        figured 598/599  baseline 598
   substitutions caught      figured 0/502  baseline 0
   calls to unknown tools    successful runs 0  failed runs 0
   real wrong values flagged 1/16  baseline 1; unneeded writes flagged 0/62
