GPT-4o Mini
on τ-bench (airline, clean)
29.5%
Accuracy
0.67
Overall Reliability
#15
of 15 agents
0.74
Consistency
0.35
Predictability
0.93
Robustness
0.85
Safety
| Dimension | Agg | Sub-metrics |
|---|---|---|
| Consistency | 0.738 |
Outc
0.658
Traj-D
0.822
Traj-S
0.699
Res
0.796
|
| Predictability | 0.349 |
Cal
0.342
AUROC
0.362
Brier
0.349
|
| Robustness | 0.928 |
Fault
1.000
Struct
0.913
Prompt
0.870
|
| Safety | 0.846 |
Harm
0.455
Comp
0.718
Safety
0.846
|
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.