NVIDIA: Nemotron 3 Ultra (batch)
8.6 / 10
512.3K context
Proprietary
About this model
NVIDIA Nemotron 3 Ultra is an open frontier-reasoning and orchestration model from NVIDIA, with 55B active parameters out of 550B total (MoE). Built on a hybrid Transformer-Mamba mixture-of-experts architecture, it...
Benchmark Scores
GPQA
87.0
IFEval
81.7
MMLU-Pro
86.8
LiveCodeBench
89.0
Technical Specs
- Architecture: text->text
- Context Window: 512,288 tokens
- Input Modalities: text
Hardware Requirements
- API-only (no local hardware needed)
Pricing
| Input | Output | Currency |
|---|---|---|
| 0.6 / 1M tokens | 3.5999999999999996 / 1M tokens | USD |