Model specification · NVIDIA
Nemotron 3 Nano 30B A3B
nvidia/nemotron-3-nano-30b-a3b
NVIDIA Nemotron 3 Nano 30B A3B is a small language MoE model with highest compute efficiency and accuracy for developers to build specialized agentic AI systems. The model is fully...
SpecificationsReported values
- Context window
- 262K
- Maximum output
- 236K
- Architecture
- text->text
- Tokenizer
- Other
- Knowledge cutoff
- —
- Moderated
- No
- Input modalities
- text
- Output modalities
- text
Full pricingUSD where reported
| Price | Standard |
|---|---|
| Input | $0.060 / M |
| Output | $0.240 / M |
Benchmark indexesReported indexes
- Intelligence
- 8.9
- Coding
- 14.4
- Agentic
- 1
- Design Arena Elo
- —
Example workload costsStandard pricing only
| Workload | Input / output | Cache share | Estimated cost |
|---|---|---|---|
| Quick check | 10K / 1K × 1 | 0% | $0.00084 |
| Long analysis | 100K / 4K × 1 | 50% | $0.00696 |
| Workday | 20K / 2K × 1,000 | 30% | $1.68 |
More from NVIDIAAll NVIDIA models
| Model | Context | Input / M | Output / M | Intel. |
|---|---|---|---|---|
| Nemotron 3 Ultra | 262K | $0.600 | $2.40 | 23.4 |
| Nemotron 3 Super | 262K | $0.080 | $0.450 | 13.6 |
| Nemotron 3.5 Lightning | 262K | $0.070 | $0.200 | 13.6 |