On October 16, 2026, Amazon Web Services announced that NVIDIA's Nemotron 3.5 Super VL multimodal foundation model is now available on Amazon SageMaker JumpStart. The new addition expands the portfolio of foundation models available for AWS customers to deploy multimodal AI solutions.
Model Capabilities and Architecture
According to Amazon, the Nemotron 3.5 Super VL vision-language model accepts images, video, and text as input and generates text. It is designed for document intelligence, GUI understanding, video question answering, and multimodal agentic applications.
The model pairs a 120B-parameter hybrid Mixture-of-Experts language backbone with 12B active parameters and a vision encoder, supporting context lengths of up to 1 million tokens. It also includes a built-in Multi-Token Prediction head for speculative decoding and supports ten languages including English, French, Spanish, German, Japanese, Hindi, Korean, and Chinese.
Deployment Options and Variants
SageMaker JumpStart offers two variants of the model. These include NVFP4 for optimized inference on G7 instances equipped with NVIDIA RTX PRO 4500 Blackwell Server Edition GPUs, and BF16 reference weights on ml.p5en.48xlarge for customization such as supervised fine-tuning, reinforcement learning, distillation, and domain adaptation.
Amazon stated that the model has been post-trained with multimodal supervised fine-tuning on image, video, document, chart, and GUI data followed by GRPO reinforcement learning, delivering structured JSON output, tool calling, and configurable reasoning.
What to do
- Navigate to the SageMaker JumpStart model catalog in the SageMaker console or use the SageMaker Python SDK to deploy the models to your AWS account.
Key facts and where they come from
- NVIDIA's Nemotron 3.5 Super VL model is now available on Amazon SageMaker JumpStart.
NVIDIA's Nemotron 3.5 Super VL is now available on Amazon SageMaker JumpStart, expanding the portfolio of foundation models available to AWS customers.
- The vision-language model accepts images, video, and text as input and generates text.
This vision-language model from the Nemotron 3.5 family accepts images, video, and text as input and generates text
- SageMaker JumpStart offers NVFP4 and BF16 reference weights variants for deployment.
SageMaker JumpStart offers two variants: NVFP4 for optimized inference on G7 instances with NVIDIA RTX PRO 4500 Blackwell Server Edition GPUs, and BF16 reference weights on ml.p5en.48xlarge
