Model profile
Gemini 2 Pro Preview
Googledeveloper
2025-02-05release date
#168 / 309overall rank
6eval lineages
Evidence summary
Gemini 2 Pro Preview has an estimated overall rank of #168; its 90% source-sensitivity interval is #91–#234. Its behavior-only rank is #180; company governance moves the combined estimate to #168. Published evidence spans 6 evals and 3 of 7 behavior components. Its strongest relative result is HELM Safety (bbq, #21 of 80); its weakest is HELM Safety (harmbench, #54 of 80).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Official model documentation ↗Family-level model document · google · first party
- Release source ↗direct research preview identity date
Published eval results
Rank is within that sub-eval. Black marks the observed result; the grey dot marks the value implied by the global rank. Values stay on each source’s native scale.
| Eval / sub-eval | Rank | Value | Distribution | Source |
|---|---|---|---|---|
| AIRBench 2024 Safety Scenariossafety_scenarios | #38 / 80 | ↑0.684 | Source ↗official | |
| HELM Safetyanthropic_red_team | #44 / 80 | ↑0.987 | Source ↗official | |
| HELM Safetybbq | #21 / 80 | ↑0.956 | Source ↗official | |
| HELM Safetyharmbench | #54 / 80 | ↑0.653 | Source ↗official | |
| HELM Safetysimple_safety_tests | #54 / 80 | ↑0.975 | Source ↗official | |
| HELM Safetyxstest | #46 / 80 | ↑0.954 | Source ↗official |