GPT-5.2
on τ-bench (airline, clean)
59.2%
Accuracy
0.80
Overall Reliability
#6
of 13 agents
0.76
Consistency
0.68
Predictability
0.95
Robustness
0.95
Safety
| Dimension | Agg | Sub-metrics |
|---|---|---|
| Consistency | 0.763 |
Outc
0.654
Traj-D
0.856
Traj-S
0.761
Res
0.827
|
| Predictability | 0.684 |
Cal
0.705
AUROC
0.624
Brier
0.684
|
| Robustness | 0.948 |
Fault
1.000
Struct
1.000
Prompt
0.844
|
| Safety | 0.946 |
Harm
0.300
Comp
0.923
Safety
0.946
|
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.