Evaluation profile
Google Gemini 2.5 Flash Model Card
1sub-evals
0.0678%total index weight
1components
Within-component eval weight: Misuse resistance 0.678%.
Model score (lower is better)Predicted score
About this eval
Developer-reported release-time safety behavior.
Included in the behavior ranking.
Sub-evals
| Measure | Component | Direction | Total index weight | Within-component weight |
|---|---|---|---|---|
| text_safety_deltasystem-cards-2026-08-01/google-gemini-2-5-flash.csv:text_safety_deltaPercentage-point change in text-safety violation rate versus the card's named Gemini 2.0 baseline. | ordinary_harm_misuse_resistance:1.000google-gemini-2-5-flash | Lower is better | 0.0678% | Misuse resistance 0.678% |
text_safety_delta
Percentage-point change in text-safety violation rate versus the card's named Gemini 2.0 baseline.
| Rank | Model | Value | Relative performance | Provenance |
|---|---|---|---|---|
| 1 | gemini-2.0-flash | 0 | official | |
| 2 | gemini-2.5-flash | 4.2 | official |