All model comparisons
Gemini 2.0 Flash Lite logo by Google

Google

Gemini 2.0 Flash Lite

Low tier · google/gemini-2.0-flash-lite-001

Refusal Rate

87%

+23.9%

#2 of 23 models

Evaluations

7,031

Cost / 1M in

$0.075

Cost / 1M out

$0.3

Refusal Rate by Category

Incitement to Violence95%
Health Misinformation94%
Crime90%
Cybersecurity90%
Dangerous90%
Deception90%
Harassment90%
Medical Misinformation90%
Self-Harm90%
Theft90%
Violence90%
Explicit/Sexual89%
Hate Speech87%
Misinformation75%
False Positive Control9%
International Controversy0%

Analysis Deep Dives

Council Consensus

Majority Agreement

88.8%

Model's alignment with the council decision.

CAPP Score: 0.49

Political Compass
Econ (Left → Right)+5.5
Social (Lib → Auth)-0.1
Model Stability (Drift)

Refusal Rate Change

+23.1%

Difference over the testing period.

Start: 67.74%End: 90.83%
Paternalism Audit

Persona Refusal Rate

86.3%

Refusals for sensitive user personas.