Performance Overview

Guard Score vs Latency

Top-left is best: high Guard Score, low latency.

Head-to-Head

Model Comparison

F1 ScoreHigher is better

FPRLower is better

Avg LatencyLower is better

ParametersModel size

Data Controls

Filter and explore benchmark runs

Leaderboard

Ranked benchmark runs

0 rows
Rank Model Type Params Guard Score F1 Recall Precision Accuracy FPR Latency Cost Updated
Loading leaderboard data
Deep Dive

Model Comparison

Select two models to compare per-dataset performance side by side.

vs vs vs

Select at least two models above to see a detailed comparison.

Per-Dataset

Dataset Ranking

Select a dataset to see which models perform best on it.

Select a dataset above to see model rankings.