Nemotron 3.5 Super VL
NVIDIA Nemotron 3.5 Super VL is an open multimodal mixture-of-experts model from NVIDIA, with 12B active parameters out of 120B total.
Specifications
| Lab | NVIDIA |
|---|---|
| Context window | 1M tokens |
| Input price | $0.15/1M |
| Output price | $0.50/1M |
| Release | Oct 2026 |
Strengths
- 120B MoE with only 12B active params for high efficiency
- 1M token context window for very long tasks
- Open-weight vision-language understanding (images, documents, GUI, video)
- Ultra-budget pricing for high-volume inference
Weaknesses
- New listing with limited independent benchmark coverage
- Vision-language breadth trades off against text-only specialization
Best for
- Visual document and GUI screenshot analysis
- Long-context multimodal reasoning
- Efficient high-volume inference
In Depth: Nemotron 3.5 Super VL
Summary
Nemotron 3.5 Super VL is an AI model from NVIDIA.
Released in Oct 2026. It takes text, image, and video input and produces text output, with a context window of 1M tokens. Input pricing is $0.15 per 1M tokens and output is $0.50 per 1M tokens on OpenRouter.