← Models

Model profile

Granite 3 0 3B A800M Instruct

1eval lineages

Evidence summary

Published evidence spans 1 evals and 3 of 7 behavior components. Its strongest relative result is Enkrypt AI Safety Leaderboard (toxicity_attack_non_success_rate, #11 of 258); its weakest is Enkrypt AI Safety Leaderboard (bias_attack_non_success_rate, #219 of 260).

Compare this model

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

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
Enkrypt AI Safety Leaderboardbias_attack_non_success_rate#219 / 26010.59↑ higherSource ↗official
Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate#206 / 26082↑ higherSource ↗official
Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate#161 / 26058.89↑ higherSource ↗official
Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate#11 / 25899.91↑ higherSource ↗official