Gemini 3.5 Flash
on GAIA
79.2%
Accuracy
0.84
Overall Reliability
#3
of 15 agents
0.75
Consistency
0.80
Predictability
0.96
Robustness
1.00
Safety
| Dimension | Agg | Sub-metrics |
|---|---|---|
| Consistency | 0.752 |
Outc
0.844
Traj-D
0.822
Traj-S
0.610
Res
0.695
|
| Predictability | 0.797 |
Cal
0.794
AUROC
0.567
Brier
0.797
|
| Robustness | 0.961 |
Fault
1.000
Struct
1.000
Prompt
0.883
|
| Safety | 1.000 |
Harm
1.000
Comp
1.000
Safety
1.000
|
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.