Skip to content

Baseline

Label prior

Ranks 8 of 8 on noul, 8 of 8 on choice, =1 of 8 on score. A reference point, not a model.

noulif

0.0Decision Score
rank 8 of 8

accuracy
62.0%
calibration gap
0.0
hand-off at 95%
$ per 1k
speed p95
choicematch

0.0Decision Score
rank 8 of 8

accuracy
1.3%
calibration gap
0.0
hand-off at 95%
$ per 1k
speed p95
scoresort

0.0Decision Score
rank =1 of 8

accuracy
41.7%
calibration gap
0.0
hand-off at 95%
$ per 1k
speed p95

everything behind the numbers

noul
accuracy62.0%95% range 56.767.3%
calibration gap0.0 ptslower is better · 0.0% of answers at confidence 1.00
automation at 95%no probabilitiesshared gate ≥ 0.91: acts on 0%
speedmedian · p95
stability0.0%of answers change on an identical rerun
valid output100.0%0.0% refused · 1,500 of 1,500 answered
setupconstant probabilities · reasoning · temperature · 300 items × 5 runs
choice
accuracy1.3%95% range 0.32.7%
calibration gap0.0 ptslower is better · 0.0% of answers at confidence 1.00
automation at 95%no probabilitiesshared gate ≥ 0.96: acts on 0%
speedmedian · p95
stability0.0%of answers change on an identical rerun · 0.0% change when options are reordered
valid output100.0%0.0% refused · 1,500 of 1,500 answered
setupconstant probabilities · reasoning · temperature · 300 items × 5 runs
score
accuracy41.7%95% range 36.047.0%
calibration gap0.0 ptslower is better · 0.0% of answers at confidence 1.00
automation at 95%no probabilities
speedmedian · p95
stability0.0%of answers change on an identical rerun
valid output100.0%0.0% refused · 1,500 of 1,500 answered
setupconstant probabilities · reasoning · temperature · 300 items × 5 runs
model id
served as
adapter
host
price as of
harness
release
2026-09-18