NVIDIABase plan
Nemotron 3.5 Super VL
NVIDIA's Nemotron 3.5 Super VL via the Vercel AI Gateway: a hybrid Mamba-2, attention, and mixture-of-experts vision-language model (120B total, 12B active) for reasoning, coding, tool use, and document understanding. Available at the base tier. Uses 1 base request per send before length multipliers. Supports image input. Does not support web search.
You always get the exact model you pick — we never silently route you to another.
Specifications
| Provider | NVIDIA |
|---|---|
| Released | 2026-10-15 |
| Intelligence | Medium |
| Speed | Fast |
| Context window | 262,144 tokens |
| Knowledge cutoff | Not officially published |
| Input price | $0.20 / 1M tokens |
| Output price | $0.50 / 1M tokens |
| Request cost | 1 base request |
| Plan tier | Base |
| Input | Text, Image |
| Output | Text |
| Features | Cached input: $0.04 / 1M tokens, 1 base request per send before length multipliers, Function calling supported, Image input supported, Served through the Vercel AI Gateway (nvidia/nemotron-3.5-super-vl), Zero data retention and no prompt training |
| Model ID | nemotron-3.5-super-vl |