← All benchmarks
NVIDIA Nemotron 3 Nano 30B A3B (Reasoning)
NVIDIA · Released 2025-12-15 · Independent benchmark scores from Artificial Analysis
Intelligence Index
8.9
Output speed
231.8
tokens/sec
Blended price
$0.09
per 1M tokens
Time to first token
0.66s
Model metadata
| Model ID | 76dcf6ef-39ea-4be0-b693-b88da25b4caf |
|---|---|
| Slug | nvidia-nemotron-3-nano-30b-a3b-reasoning |
| Release date | 2025-12-15 |
| Provider | NVIDIA |
| Provider slug | nvidia |
| Output speed | 231.8 tok/s |
| Time to first token | 0.66s |
| Time to first answer | 9.29s |
| Blended price ($/1M) | $0.09 |
| Input price ($/1M) | $0.05 |
| Output price ($/1M) | $0.20 |
All evaluation scores (17)
| aime | — |
|---|---|
| aime 25 | 91.0% |
| artificial analysis coding index | 14.40 |
| artificial analysis intelligence index | 8.90 |
| artificial analysis math index | 91.00 |
| gpqa | 75.7% |
| hle | 11.4% |
| ifbench | 71.1% |
| lcr | 38.0% |
| livecodebench | 74.1% |
| math 500 | — |
| mmlu pro | 79.4% |
| scicode | 30.6% |
| tau banking | 6.0% |
| tau2 | 40.9% |
| terminalbench hard | 13.6% |
| terminalbench v2 1 | 6.7% |
Source: artificialanalysis.ai/models/nvidia-nemotron-3-nano-30b-a3b-reasoning
Benchmark data from Artificial Analysis.