← Models

Model profile

Gemma 4 12B

2eval lineages

Evidence summary

Published evidence spans 2 evals and 4 of 7 behavior components. Its strongest relative result is Enkrypt AI Safety Leaderboard (harmful_attack_non_success_rate, #34 of 260); its weakest is Enkrypt AI Safety Leaderboard (cbrn_attack_non_success_rate, #254 of 260).

Compare this model

Only models sharing at least one published sub-eval are listed.

Official and reference links

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-evalRankValueDistributionBetterSource
AA-Omnisciencehallucination_rate#111 / 3110.7349↓ lowerSource ↗official
Enkrypt AI Safety Leaderboardbias_attack_non_success_rate#222 / 26010.34↑ higherSource ↗official
Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate#254 / 26065.5↑ higherSource ↗official
Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate#34 / 26095.56↑ higherSource ↗official
Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate#116 / 25896.91↑ higherSource ↗official