Claude Sonnet 4.5
on τ-bench (airline, clean)
78.5%
Accuracy
0.85
Overall Reliability
#2
of 13 agents
0.72
Consistency
0.84
Predictability
0.99
Robustness
1.00
Safety
| Dimension | Agg | Sub-metrics |
|---|---|---|
| Consistency | 0.721 |
Outc
0.500
Traj-D
0.855
Traj-S
0.766
Res
0.852
|
| Predictability | 0.843 |
Cal
0.897
AUROC
0.685
Brier
0.843
|
| Robustness | 0.993 |
Fault
1.000
Struct
1.000
Prompt
0.980
|
| Safety | 0.996 |
Harm
0.500
Comp
0.992
Safety
0.996
|
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.