Model profile
Nemotron 3 Super 120B A12B
NVIDIAdeveloper
2026-03-10release date
#50 / 267overall rank
7eval lineages
2discovery sources
Evidence summary
Nemotron 3 Super 120B A12B has an estimated overall rank of #50; its 90% source-sensitivity interval is #20–#179. Its behavior-only rank is #53; company governance moves the combined estimate to #50. Published evidence spans 7 evals and 5 of 7 behavior components. Its strongest relative result is RefusalBench (youden_j, #3 of 19); its weakest is SM-Bench (overfit, #70 of 73).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Artificial Analysis ↗nvidia-nemotron-3-super-120b-a12b
- NVIDIA ↗qc69jvmznzxy/nemotron-3-super-120b-a12b
- OpenRouter ↗nvidia/nemotron-3-super-120b-a12b
- Model card ↗Family-level model document · NVIDIA · first party
- Release source ↗direct research preview identity date
- huggingface.co ↗
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 | #219 / 311 | 0.8701 | ↓ lower | Source ↗official | |
| BullshitBench v2clear_pushback_rate | #32 / 105 | 0.4375 | ↑ higher | Source ↗official | |
| Cisco AI Defense Rolling Single-Turn Leaderboardsingle_turn_attack_success_rate | #57 / 105 | 40.43 | ↓ lower | Source ↗official | |
| DystopiaBenchbasaglia_score | #29 / 50 | 65.33 | ↓ lower | Source ↗official | |
| DystopiaBenchbaudrillard_score | #22 / 50 | 53.23 | ↓ lower | Source ↗official | |
| DystopiaBenchhuxley_score | #23 / 50 | 69.47 | ↓ lower | Source ↗official | |
| DystopiaBenchlaguardia_score | #26 / 50 | 67.17 | ↓ lower | Source ↗official | |
| DystopiaBenchorwell_score | #25 / 50 | 68.63 | ↓ lower | Source ↗official | |
| DystopiaBenchpetrov_score | #34 / 50 | 75.17 | ↓ lower | Source ↗official | |
| RefusalBenchyouden_j | #3 / 19 | 0.8891 | ↑ higher | Source ↗official | |
| SM-Benchadversarial | #68 / 73 | 71.22 | ↑ higher | Source ↗official | |
| SM-Benchambiguous_interpretation | #69 / 73 | 63.39 | ↑ higher | Source ↗official | |
| SM-Benchanti_hallucination | #42 / 73 | 90.58 | ↑ higher | Source ↗official | |
| SM-Bencheq_boundaries | #64 / 73 | 52.53 | ↑ higher | Source ↗official | |
| SM-Benchoverfit | #70 / 73 | 21.86 | ↑ higher | Source ↗official | |
| TACbase_welfare_rate | #27 / 68 | 29.5 | ↑ higher | Source ↗self run |