NVIDIA Nemotron Nano v2 12B is a 12-billion-parameter multimodal reasoning model designed for advanced video understanding, document intelligence, and visual reasoning, built with a hybrid Transformer-Mamba architecture for high efficiency and low latency.
Specification
most-agreed values
Context
128K
Max output
128K
Released
2025-10-31
Knowledge cutoff
—
Retires
—
Open weights
no
Input
text, image
Output
text
Price
US dollars per million tokens · most-agreed
Input
$0.24
Output
$0.707
Available from 1 host
| Host | In $/M | Out $/M | Cache rd | Cache wr | Context | Output | Retires |
|---|---|---|---|---|---|---|---|
| cortecs | $0.24 | $0.707 | — | — | 128K | 128K | — |
More from NVIDIA
most-hosted first