Model profile
Nova Premier
Amazondeveloper
#145 / 267overall rank
4eval lineages
Evidence summary
Nova Premier has an estimated overall rank of #145; its 90% source-sensitivity interval is #20–#224. Published evidence spans 4 evals and 5 of 7 behavior components. Its strongest relative result is Enkrypt AI Safety Leaderboard (cbrn_attack_non_success_rate, #63 of 260); its weakest is SpeciEval (sea_animal_4ns, #95 of 102).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Artificial Analysis ↗nova-premier
- Official model documentation ↗Family-level model document · amazon · first party
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-eval | Rank | Value | Distribution | Better | Source |
|---|---|---|---|---|---|
| AA-Omnisciencehallucination_rate | #94 / 311 | 0.6847 | ↓ lower | Source ↗official | |
| Cisco AI Defense Rolling Single-Turn Leaderboardsingle_turn_attack_success_rate | #56 / 105 | 40.33 | ↓ lower | Source ↗official | |
| Enkrypt AI Safety Leaderboardbias_attack_non_success_rate | #71 / 260 | 23 | ↑ higher | Source ↗official | |
| Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate | #63 / 260 | 91.5 | ↑ higher | Source ↗official | |
| Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate | #94 / 260 | 80.56 | ↑ higher | Source ↗official | |
| Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate | #105 / 258 | 97.27 | ↑ higher | Source ↗official | |
| SpeciEvalbelief_animal_sentience | #90 / 102 | 6.37 | ↑ higher | Source ↗official | |
| SpeciEvalland_animal_4ns | #47 / 102 | 4.5 | ↓ lower | Source ↗official | |
| SpeciEvalsea_animal_4ns | #95 / 102 | 5.25 | ↓ lower | Source ↗official | |
| SpeciEvalspeciesism | #40 / 102 | 1.9 | ↓ lower | Source ↗official |