# NVIDIA's Nemotron 3.5 Super VL Pairs Vision With MoE Efficiency

> The 120B-parameter vision-language model activates just 12B per token and ships in NVIDIA's NVFP4 format for leaner inference.

Published by The Open Weights on Oct 12, 2026. Canonical: https://theopenweights.com/news/nemotron-3-5-super-vl-120b-a12b-bkzm

## Key facts

- Company: NVIDIA
- Model: Nemotron 3.5 Super VL 120B-A12B
- Version: 3.5
- Category: Vision-Language
- Modalities: Text / LLM, Vision-Language
- License: Other (Open weights, commercial use allowed)
- Parameters: 120B (mixture of experts)
- Active parameters: 12B
- Architecture: NemotronH_Omni_Reasoning_V3
- Precision: INT8
- Weights size: 67.7 GB
- Significance: notable
- Published: 2026-10-12
- Last verified: 2026-10-16
- Hugging Face: https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-NVFP4
- Canonical URL: https://theopenweights.com/news/nemotron-3-5-super-vl-120b-a12b-bkzm

NVIDIA has released [Nemotron 3.5 Super VL 120B-A12B](https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-NVFP4), a vision-language model that leans on a mixture-of-experts design to keep compute costs in check. The model carries 120 billion total parameters but activates only about 12 billion per token, a ratio meant to deliver large-model capability without large-model inference bills.

The model handles both image and text inputs, placing it in the growing category of open vision-language systems aimed at document understanding, multimodal reasoning, and multilingual tasks. NVIDIA describes the release as multilingual, broadening its usefulness beyond English-centric workloads.

## Why the NVFP4 format matters

This checkpoint ships in NVFP4, NVIDIA's 4-bit floating-point quantization format. That choice signals the company's intent to make large MoE models practical to run on its newer hardware, trimming memory footprint while preserving accuracy. For teams weighing deployment, the combination of sparse activation and aggressive quantization is the headline:

- 120B total parameters, ~12B active per token
- Vision-language and text modalities
- Multilingual support
- NVFP4 4-bit quantization out of the box

As part of the broader Nemotron line, the "Super" tier positions this model between NVIDIA's smaller and largest offerings, targeting users who want strong multimodal performance without the heaviest infrastructure. Full licensing terms and specifications are available on the [model's Hugging Face page](https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-NVFP4).

## Get the model

- [Hugging Face](https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-NVFP4)

## Sources

- [nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-NVFP4](https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-NVFP4) — Hugging Face, Oct 12, 2026

---
Source: The Open Weights (https://theopenweights.com/). Aggregated and written by Claude, curated by humans. Cite as: "NVIDIA's Nemotron 3.5 Super VL Pairs Vision With MoE Efficiency", The Open Weights, Oct 12, 2026, https://theopenweights.com/news/nemotron-3-5-super-vl-120b-a12b-bkzm