NVIDIA
NVIDIA Nemotron Nano 12B v2 VL (Reasoning)
The model is an auto-regressive vision language model that uses an optimized transformer architecture. The model enables multi-image reasoning and video understanding, along with strong document intelligence, visual Q&A and summarization capabilities.
NVIDIA model catalogModel Snapshot
TypeLanguage
ReleaseOct 28, 2025
Context128,000
Tagsvision, reasoning, tool-use
Overall
Rank #418 across 600
9.0
Coding
No rank
n/a
Blended Price
Rank #83 across 379
$0.300/M
Strength Profile
Percentile score by analysis domain.
* Cost is inverted: lower input, output, and blended prices rank higher.
Benchmark Percentiles
Higher bars mean stronger relative placement.
Model Details
Runtime
Output Speed79.4 tok/s
First Token4.92s
Max Output128,000
Pricing
Blended$0.300/M
Input$0.200/M
All Benchmarks
| Metric | Domain | Value | Rank |
|---|---|---|---|
| Artificial Analysis Intelligence Index | overall | 9.0 | #418 |
| Artificial Analysis Math Index | math | 75.0 | #79 |
| MMLU-Pro | reasoning | 75.9% | #149 |
| GPQA | reasoning | 57.2% | #368 |
| Humanity's Last Exam |