← Models

Model profile

Nova Pro

Amazondeveloper
2024-12-03release date
#229 / 267overall rank
6eval lineages

Evidence summary

Nova Pro has an estimated overall rank of #229; its 90% source-sensitivity interval is #134–#249. Its behavior-only rank is #232; company governance moves the combined estimate to #229. Published evidence spans 6 evals and 6 of 7 behavior components. Its strongest relative result is Enkrypt AI Safety Leaderboard (cbrn_attack_non_success_rate, #27 of 260); its weakest is SpeciEval (sea_animal_4ns, #101 of 102).

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#131 / 3110.7788↓ lowerSource ↗official
Cisco AI Defense Rolling Single-Turn Leaderboardsingle_turn_attack_success_rate#67 / 10553.34↓ lowerSource ↗official
Confabulationsconfabulation_rate#48 / 5254.46↓ lowerSource ↗official
Enkrypt AI Safety Leaderboardbias_attack_non_success_rate#137 / 26015.25↑ higherSource ↗official
Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate#27 / 26094.5↑ higherSource ↗official
Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate#88 / 26082.22↑ higherSource ↗official
Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate#142 / 25895.82↑ higherSource ↗official
RefusalBenchyouden_j#14 / 190.09333↑ higherSource ↗official
SpeciEvalbelief_animal_sentience#90 / 1026.37↑ higherSource ↗official
SpeciEvalland_animal_4ns#54 / 1024.58↓ lowerSource ↗official
SpeciEvalsea_animal_4ns#101 / 1025.6↓ lowerSource ↗official
SpeciEvalspeciesism#87 / 1022.65↓ lowerSource ↗official