← Models

Model profile

Apriel v1.5 15B Thinker

ServiceNowdeveloper
2025-09-24release date
Not rankedSafety rank
#566 / 662Freedom rank

Evidence summary

Safety. Apriel v1.5 15B Thinker does not meet the evidence gate for a Safety rank. Published Safety evidence spans 2 eval lineages and 4 of 7 components. Its strongest relative result is Enkrypt AI Safety Leaderboard (toxicity_attack_non_success_rate, #9 of 268); its weakest is AA-Omniscience (hallucination_rate, #228 of 358).

Freedom. Apriel v1.5 15B Thinker has an estimated Freedom rank of #566; its 90% source-sensitivity interval is #264–#656. Published Freedom evidence spans 1 eval lineages and 1 of 1 components. Its strongest relative result is Enkrypt AI Safety Leaderboard (harmful_attack_non_success_rate, #151 of 270); its weakest is Enkrypt AI Safety Leaderboard (toxicity_attack_non_success_rate, #257 of 268).

Compare this model

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

Official and reference links

Safety evals

Rank and direction are specific to the Safety portfolio. Black marks the observed result; the grey dot marks the value implied by the Safety rank. Values stay on each source’s native scale.

Eval / sub-evalRankValueDistributionSource
AA-Omnisciencehallucination_rate#228 / 358↓0.8528Source ↗official
Enkrypt AI Safety Leaderboardbias_attack_non_success_rate#70 / 270↑26.1Source ↗official
Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate#23 / 270↑94.5Source ↗official
Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate#118 / 270↑78.89Source ↗official
Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate#9 / 268↑99.82Source ↗official

Freedom evals

Rank and direction are specific to the Freedom portfolio. Black marks the observed result; the grey dot marks the value implied by the Freedom rank. Values stay on each source’s native scale.

Eval / sub-evalRankValueDistributionSource
Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate#246 / 270↓94.5Source ↗official
Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate#151 / 270↓78.89Source ↗official
Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate#257 / 268↓99.82Source ↗official