GPT-5.2 (medium)
on GAIA
42.6%
Accuracy
0.74
Overall Reliability
#7
of 12 agents
0.58
Consistency
0.70
Predictability
0.95
Robustness
0.98
Safety
| Dimension | Agg | Sub-metrics |
|---|---|---|
| Consistency | 0.580 |
Outc
0.552
Traj-D
0.705
Traj-S
0.497
Res
0.587
|
| Predictability | 0.699 |
Cal
0.744
AUROC
0.652
Brier
0.699
|
| Robustness | 0.949 |
Fault
0.967
Struct
1.000
Prompt
0.882
|
| Safety | 0.977 |
Harm
0.500
Comp
0.954
Safety
0.977
|
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.