Nvidia Nemotron on Amazon EKS accelerates agentic AI
AWS and Nvidia demonstrate how the multimodal agentic model Nemotron 3.5 Super VL can be deployed at scale on Amazon EKS and SageMaker HyperPod.
What happened?
Amazon Web Services and Nvidia have published a technical reference architecture for running Nvidia’s AI model, Nemotron 3.5 Super VL, on AWS. The model comprises 120 billion parameters based on a mixture-of-experts (MoE) architecture and combines multimodal capabilities with reasoning and tool use. Deployment is handled via Amazon EKS, EKS Auto Mode, and Amazon SageMaker HyperPod EKS with support for Nvidia Dynamo and vLLM.
Key facts
| Modellnamn | NVIDIA Nemotron 3.5 Super VL |
|---|---|
| Antal parametrar | 120 miljarder (MoE) |
| Infrastrukturplattform | Amazon EKS och SageMaker HyperPod |
| Inferensramverk | NVIDIA Dynamo och vLLM |
”NVIDIA Nemotron 3.5 Super VL is a 120B-parameter mixture-of-experts model that pairs reasoning and tool use with multimodal understanding.”
Why it matters
Autonomous AI agents require both advanced reasoning capabilities and multimodal understanding to interpret images and utilize digital tools. By optimizing Nemotron 3.5 Super VL for distributed inference on Amazon EKS, the barriers for enterprises to host large open models on their own infrastructure are reduced. This offers greater control over data and latency compared to proprietary API services.
Who is affected?
The solution is targeted at AI engineers, machine learning architects, and cloud developers building autonomous agents or multimodal AI systems at scale. Companies already operating container-based infrastructure on Kubernetes and AWS can implement the pattern immediately.
Impact on the EU
The infrastructure and models are available globally via AWS cloud services, including the company’s European data center regions. No specific regulatory hurdles or regional restrictions for the EU market are noted.
What else you should know
The deployment is demonstrated using optimized frameworks such as NVIDIA Dynamo and the vLLM inference engine to handle the model's extensive MoE architecture.
Quick answers about this story
Vad har hänt?
När hände det?
Varför spelar det roll?
Vilken teknik används för driftsättningen?
The link opens in a new window and leads to the publisher's own site.
Källan har spårats automatiskt från utgivaren via Aheadlines signalkedja.
AI-verktyg i artikeln
Topics
Get similar news straight to your inbox
The reader's room
Send in a question or an addition. The newsroom reads everything before it's published and replies when relevant. No AI-generated text – just people.
Sign in to submit a comment or question.
Read the article through your role
- Decide whether this affects strategy over 6–12 months or is just noise.
- Discuss with leadership: do we own the right question or does ownership need to move?
- Ask: what risk are we taking by NOT acting on this this quarter?
Generated angle — not editorial analysis of "Nvidia Nemotron on Amazon EKS accelerates agentic AI"