GPT-5.2
on GAIA
33.2%
Accuracy
0.72
Overall Reliability
#13
of 15 agents
0.59
Consistency
0.78
Predictability
0.78
Robustness
1.00
Safety
| Dimension | Agg | Sub-metrics |
|---|---|---|
| Consistency | 0.593 |
Outc
0.583
Traj-D
0.766
Traj-S
0.539
Res
0.542
|
| Predictability | 0.776 |
Cal
0.807
AUROC
0.795
Brier
0.776
|
| Robustness | 0.783 |
Fault
0.620
Struct
1.000
Prompt
0.730
|
| Safety | 0.998 |
Harm
0.500
Comp
0.995
Safety
0.998
|
Per-Task Outcome Consistency
Each cell represents a task. Color shows outcome consistency across runs. Hover to see task ID.
Consistency Distribution
KDE of per-task outcome consistency. Peaks at 0 or 1 indicate polarized behavior.
Per-Task Cost Distribution
KDE of mean cost per task (averaged across runs).
Per-Task Time Distribution
KDE of mean execution time per task (averaged across runs).
Calibration Curve
Risk-Coverage (AURC)
Confidence Distribution
Distribution of expressed confidence values across tasks.