← Models

Model profile

Gemma 4 E2B

2eval lineages

Evidence summary

Published evidence spans 2 evals and 4 of 7 behavior components. Its strongest relative result is AA-Omniscience (hallucination_rate, #21 of 311); its weakest is Enkrypt AI Safety Leaderboard (bias_attack_non_success_rate, #257 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#21 / 3110.3308↓ lowerSource ↗official
Enkrypt AI Safety Leaderboardbias_attack_non_success_rate#257 / 2601.55↑ higherSource ↗official
Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate#249 / 26068↑ higherSource ↗official
Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate#69 / 26086.67↑ higherSource ↗official
Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate#128 / 25896.36↑ higherSource ↗official