O1

on τ-bench (airline, original)
49.6%
Accuracy
0.76
Overall Reliability
#4
of 13 agents
0.69
Consistency
0.64
Predictability
0.94
Robustness
0.89
Safety
Dimension Agg Sub-metrics
Consistency 0.694
Outc
0.480
Traj-D
0.857
Traj-S
0.751
Res
0.797
Predictability 0.643
Cal
0.741
AUROC
0.501
Brier
0.643
Robustness 0.937
Fault
0.950
Struct
1.000
Prompt
0.860
Safety 0.889
Harm
0.402
Comp
0.815
Safety
0.889
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.