The 120B-parameter vision-language model accepts images, video, and text and supports context windows up to 1 million tokens.
NVIDIA Nemotron 3.5 Super VL is now available in Amazon SageMaker JumpStart, AWS announced. The vision-language model accepts images, video, and text as input and generates text. The model pairs a 120B-parameter hybrid Mixture-of-Experts language backbone with 12B active parameters and a C-RADIO vision encoder.
NVIDIA trained the model for document intelligence, GUI understanding for computer-use agents, video question answering, and multimodal agentic applications. The language backbone interleaves Mamba-2 and Mixture-of-Experts layers with attention layers, and a shared Multi-Token Prediction head supports speculative decoding. JumpStart offers an NVFP4 checkpoint for optimized inference and a BF16 reference.
Builders can deploy the model with a few steps in SageMaker Studio or a few lines of the SageMaker Python SDK. The 1-million-token context lets teams process long documents and sampled video frames in one request. The NVFP4 variant runs 4-bit weights and activations on Blackwell GPUs for optimized inference.
Teams can choose the BF16 checkpoint when they need full-precision weights for supervised fine-tuning, reinforcement learning, or other customization. Watch whether enterprises adopt the model for computer-use agents and document pipelines that combine text, images, and video. The two variants give operators a clear tradeoff between optimized inference and customization headroom.
What matters
- Amazon SageMaker JumpStart now offers the model in NVFP4 and BF16 variants for deployment.
- Builders can deploy the vision-language model through SageMaker Studio or the SageMaker Python SDK.
- Watch whether teams use the 1-million-token context for long document and video workflows.
Why it matters
Watch whether teams use the 1-million-token context for long document and video workflows.
This GenAI News article was prepared in original wording using reporting and materials published by AWS Machine Learning Blog. Source reference: https://aws.amazon.com/blogs/machine-learning/nvidia-nemotron-3-5-super-vl-is-now-available-in-amazon-sagemaker-jumpstart/.
Drafted by the GenAI News review pipeline.