Your sovereign, private NVIDIA Nemotron Super 3 120B
Nemotron Super 3 120B is part of NVIDIA's Nemotron family of open models with open weights, training data and recipes: a frontier-scale hybrid Mamba-2+MoE+Attention reasoner built for agentic workflows and efficient long-context inference. A reasoning model in the extra-large tier, served on ToothFairyAI's AU, EU and US endpoints — private by default, billed in US dollars per million tokens at the published rates below. No training on your data, no vendor lock-in.
How NVIDIA Nemotron Super 3 120B compares
Independent intelligence index for NVIDIA Nemotron Super 3 120B against every peer in the Extra-large tier (and the tier above where needed) — same source, same tests.
Latency — end-to-end seconds for a 500-token answer
Capability indexes
Domain-weighted agentic performance for NVIDIA Nemotron Super 3 120B.
Capability index vs extra-large peers
The evaluations behind the index
Independent benchmark results for NVIDIA Nemotron Super 3 120B — Elo scores as published; pass rates shown as percentages.
Per-evaluation economics
| Evaluation | Score | Time / task | Output tokens / task |
|---|---|---|---|
| AA-Briefcase | 0.0% | 56.9 min | 604419 |
| GDPval-AA | 477.33 | 2.3 min | 24750 |
| AutomationBench-AA | 3.8% | 1.2 min | 12246 |
| Terminal-Bench 4.0 | 0.0% | 8.3 min | 87694 |
| SciCode | 36.2% | 0.2 min | 1915 |
| Humanity's Last Exam | 20.8% | 3.4 min | 36199 |
| GDP.pdf | 2.6% | 0.7 min | 7700 |
| CritPt | 3.1% | 5.6 min | 59305 |
| AA-Omniscience | -4150.0% | 0.2 min | 2274 |
| AA-LCR (long context) | 65.7% | 0.8 min | 8998 |
Run NVIDIA Nemotron Super 3 120B privately
One key, one endpoint, three jurisdictions.
- Size tierExtra-large
- Serving routeStandard
- Input rate$1.56/1M
- Output rate$4.68/1M
- Cached input rate—
- Context window164K tokens
- Max output16K tokens
- ReasoningYes — emits thinking traces
- Vision & video inputText only
- Tool callingSupported
- Data residencyAU, EU and US endpoints — processing stays inside the region you call