NVIDIA Nemotron 3.5 Super VL 120B A12B Base BF16
nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-Base-BF16
Open Source · chat · open-weights
Open
Alert me on changes
Context
1M
Max output
8.2K
Weights
Open
API $/1M
—
Modalities
text · image
Released
12 Oct 2026
License: other · nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-Base-BF16
AI summary
● machine-written
NVIDIA Nemotron 3.5 Super VL 120B: Open multimodal MoE model
NVIDIA Nemotron 3.5 Super VL is an open-weights vision-language mixture-of-experts model with 120B total parameters and 12B active parameters. It supports 1M token context length and handles text, image, and video inputs for reasoning and agentic tasks. The model is available through multiple providers with varying pricing and deployment options.
What's new
- 120B total parameters with 12B active parameters via mixture-of-experts architecture
- Supports 1M token context window for long-document reasoning
- Native Multi-Token Prediction (MTP) enables speculative decoding
- Available open-weights for commercial or non-commercial use
- Supports text, image, and video modalities
Best for
Multi-agent applications and complex reasoning tasksLong-context document analysis and cross-document reasoningVision-language understanding and agentic tool useMulti-step task planning
Source: https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-Base-BF16