← All benchmarks
NVIDIA Nemotron Nano 12B v2 VL (Reasoning)
NVIDIA · Released 2025-10-28 · Independent benchmark scores from Artificial Analysis
Intelligence Index
9.0
Output speed
77.0
tokens/sec
Blended price
$0.30
per 1M tokens
Time to first token
4.60s
Model metadata
| Model ID | 6e6e02fd-9cbd-417f-9bfc-673df89c313d |
|---|---|
| Slug | nvidia-nemotron-nano-12b-v2-vl-reasoning |
| Release date | 2025-10-28 |
| Provider | NVIDIA |
| Provider slug | nvidia |
| Output speed | 77.0 tok/s |
| Time to first token | 4.60s |
| Time to first answer | 30.57s |
| Blended price ($/1M) | $0.30 |
| Input price ($/1M) | $0.20 |
| Output price ($/1M) | $0.60 |
All evaluation scores (17)
| aime | — |
|---|---|
| aime 25 | 75.0% |
| artificial analysis coding index | — |
| artificial analysis intelligence index | 9.00 |
| artificial analysis math index | 75.00 |
| gpqa | 57.2% |
| hle | 5.3% |
| ifbench | 31.9% |
| lcr | 40.0% |
| livecodebench | 69.4% |
| math 500 | — |
| mmlu pro | 75.9% |
| scicode | 26.2% |
| tau banking | — |
| tau2 | 21.3% |
| terminalbench hard | 4.5% |
| terminalbench v2 1 | — |
Source: artificialanalysis.ai/models/nvidia-nemotron-nano-12b-v2-vl-reasoning
Benchmark data from Artificial Analysis.