AWS lists Nemotron 3.5 Super VL on EKS and SageMaker HyperPod EKS
Amazon Web Services now lists NVIDIA Nemotron 3.5 Super VL on Amazon EKS and SageMaker HyperPod EKS alongside existing Nemotron 3 Nano, Nano Omni, Super, Ultra, and Nemotron 3.5 Lightning offerings on AWS. AWS describes a 120-billion-parameter mixture-of-experts model activating twelve billion parameters per token, combining reasoning, tool use, and multimodal understanding for text, images, documents, and video in agentic workloads. For teams on EKS or HyperPod, the update adds a vendor-supported vision-language MoE path without changing cloud provider. NVIDIA Research feed excerpts on related Nemotron 3.5 Super and Lightning hybrids supply titles only, without benchmarks or licensing facts, so on-AWS performance and economics remain unevaluated in this packet.
AWS lists Nemotron 3.5 Super VL on EKS and SageMaker HyperPod EKS
NVIDIA Nemotron 3.5 Super VL is now offered alongside other Nemotron 3 variants already on AWS. AWS describes it as a 120B-parameter MoE model with 12B parameters active per token, combining reasoning, tool use, and text, image, document, and video understanding for agentic workloads on EKS and HyperPod EKS.
Key takeaway
AWS operators can deploy Nemotron 3.5 Super VL on EKS and HyperPod EKS as a listed 120B MoE multimodal agent model with 12B active parameters per token.
What happened
According to AWS ML blog coverage summarized in the packet, NVIDIA Nemotron 3.5 Super VL joins other Nemotron 3 family models already available on AWS and is offered for agentic AI on Amazon EKS and SageMaker HyperPod EKS.
AWS attributes to the model a 120B-parameter mixture-of-experts architecture with 12B parameters activated per token, plus combined reasoning, tool use, and understanding of text, images, documents, and video for agent workloads on those platforms.
Evidence
Nemotron 3.5 Super VL is listed on Amazon EKS and SageMaker HyperPod EKS on AWS.
AWS ML · attributed
NVIDIA Nemotron 3.5 Super VL joins the NVIDIA Nemotron 3 Nano , Nano Omni , Super , Ultra , and Nemotron 3.5 Lightning models already available on AWS.
AWS describes Nemotron 3.5 Super VL as a 120B MoE model with 12B active parameters per token and multimodal agent capabilities.
AWS ML · attributed
It's a 120B-parameter mixture-of-experts (MoE) model that activates 12B parameters per token. It combines reasoning and tool use with text, image, document, and video understanding
NVIDIA Research feed material frames Nemotron 3.5 Super and Lightning as open, efficient MoE hybrid Mamba-Transformer models for agentic AI at title level only.
NVIDIA Research Official · attributed
Nemotron 3.5 Super and Lightning: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Models for Agentic AI - NVIDIA
The NVIDIA-related excerpt in the packet does not establish benchmarks, parameter counts, licensing terms, or availability details.
NVIDIA Research Official · attributed
The excerpt provides only the title and a one-line description, so no benchmark results, parameter counts, licensing terms, or availability details are established.
Why it matters
Listing Super VL on managed AWS Kubernetes surfaces gives builders a consolidated cloud path for multimodal agent stacks, while the thin NVIDIA excerpt limits how far efficiency or openness claims can be weighed from this packet alone.
Limits and uncertainties
The AWS blog excerpt in the packet is truncated mid-sentence after describing multimodal understanding, so downstream deployment steps and limits are not fully visible here.
NVIDIA Research coverage in the feed is excerpt-only, with no benchmark results, licensing terms, or availability details established in the packet.
Practical implications
Teams on EKS or HyperPod EKS can evaluate Nemotron 3.5 Super VL against other Nemotron 3 variants already on AWS when designing agentic pipelines that need vision-language and tool use in one model slot.
Operators should treat NVIDIA open and efficiency framing for related Nemotron 3.5 Super and Lightning variants as unverified until fuller primary sources or AWS run metrics are available.
What to watch
Whether AWS publishes complete HyperPod EKS and EKS deployment guidance, pricing signals, or operational limits for Nemotron 3.5 Super VL beyond the truncated excerpt.
Whether NVIDIA releases primary documentation with benchmarks, licensing, and availability that substantiate agentic efficiency claims for Nemotron 3.5 Super and Lightning.
