94GRD90ADH89SYN90PRC85EFF92CON
83–98 across 4 of 5 boards
confirmation-desk-dayRun #133—risk-limit-breach-dayRun #10198#1high-board-portfolio-review-dayRun #9483#7trader-rfq-booking-dayRun #3391#1risk-manager-control-dayRun #2086#1
Every ability card measured for gpt-5-6-terra.
averaged across the boards this model contested
83–98 across 4 of 5 boards
confirmation-desk-dayRun #133—risk-limit-breach-dayRun #10198#1high-board-portfolio-review-dayRun #9483#7trader-rfq-booking-dayRun #3391#1risk-manager-control-dayRun #2086#14 measurements · each against that board's own field
risk-limit-breach-dayRun #101
obj 97.4 · 2 trials
high-board-portfolio-review-dayRun #94
obj 82.9 · 2 trials
trader-rfq-booking-dayRun #33
obj 96.0 · 2 trials
risk-manager-control-dayRun #20
obj 89.8 · 2 trials