Build Visual AI Agents With NVIDIA Nemotron 3.5 Super VL @NVIDIADeveloper
Build Visual AI Agents With NVIDIA Nemotron 3.5 Super VL  @NVIDIADeveloper
Uploaded October 2026 | Updated November 2026, 4 days ago
Learn how NVIDIA Nemotron 3.5 Super VL investigates why an AI bill rose 30% using a dashboard screenshot and a scanned pricing sheet.

In this walkthrough, Chris Alexiuk introduces the open 120-billion-parameter mixture-of-experts model with 12 billion active parameters, then Chia-Chih Chen demonstrates a visual investigation agent. The agent reads the dashboard, identifies a surge in thinking-token use, writes Python to compare routing options and prepares a proposal for human approval. The recommended plan stays within the team’s quality bar while projecting about $100,000 in monthly savings.

NVIDIA Nemotron 3.5 Super VL combines multimodal reasoning, tool use and code generation for agents that work across text, images, charts, documents and video. It can also run on a single NVIDIA DGX Spark.

Use the links and chapters below to explore the model, technical blog and developer resources.

➑️ Learn more: blogs.nvidia.com/blog/nemotron-3-5-super-vl
πŸ“ Technical blog: developer.nvidia.com/blog/nvidia-nemotron-3-5-super-vl-powers-specialized-agents-with-multimodal-reasoning
πŸ“₯ Download the model: huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-Base-BF16
πŸš€ Try on NVIDIA Build: build.nvidia.com/nvidia/nemotron-3.5-super-vl-120b-a12b
πŸ“š Super VL cookbook: github.com/NVIDIA-NeMo/Nemotron/tree/main/usage-cookbook/Nemotron-3.5-Super-VL/visual-investigation-agent
πŸ“„ Technical report: research.nvidia.com/labs/nemotron/files/NVIDIA-Nemotron-3.5-Super-and-Lightning-Technical-Report.pdf

Chapters:
00:00 – Introduction
00:07 – NVIDIA Nemotron 3.5 Super VL overview
00:31 – Vision-language capabilities
01:09 – Multimodal billing investigation demo
01:46 – Read dashboard metrics
01:54 – Parse a scanned pricing sheet
02:01 – Find the cause of the cost increase
02:09 – Optimize ticket routing with Python
02:16 – Prepare a proposal for human review
02:28 – Cookbook and developer resources
Build Visual AI Agents With NVIDIA Nemotron 3.5 Super VLDassault Systèmes Accelerates Simulation with NVIDIA AI & GPUsNemotron 3.5 Lightning Q&A: Sub-agents, vLLM or Ollama?Missed our Cosmos 3 livestream? Weve got you covered.What 5,000 Kagglers Taught Us About Improving AI Reasoning | Nemotron LabsGenerating Synthetic Defect Images for Visual Inspection With Open Source Physical AI Agent SkillsIntroducing NVIDIA Nemotron 3.5 Super VLA Differential Monte Carlo Solver For the Poisson Equation | NVIDIA ResearchGTC Dev Community Berlin Recap: From Open Models to Physical AIServing World Models: Latency vs Throughput with NVIDIA Cosmos 3 Super on Nebius | Cosmos LabsA chess grandmaster entered our hackathon. Heres what he built.Ask the Experts: How NVIDIA OpenShell Secures Autonomous Agents | Nemotron Labs
NVIDIA Developer |

Build Visual AI Agents With NVIDIA Nemotron 3.5 Super VL

SHARE TO X SHARE TO REDDIT SHARE TO FACEBOOK WALLPAPER