Skip to main content
← Back to sources

NVIDIA Demonstrates Gemma 4 VLA Model Running on Jetson Orin Nano Super

Published 2026-04-23AI Infrastructure and ComputeMedium

Summary

NVIDIA has published a demonstration of Google's Gemma 4 Vision-Language-Action (VLA) model running on the Jetson Orin Nano Super, a compact edge AI computing platform. The blog post, authored by Asier Arranz and hosted on Hugging Face's platform, showcases the feasibility of deploying multimodal foundation models for robotics and embodied AI tasks on edge hardware. VLA models combine vision, language understanding, and action generation, making them significant for autonomous robotics applicati

Alignment: Reinforces current position
Related Positions: ai-infrastructure-strategy.md, multi-model-multi-vendor.md
nvidiagemma-4jetson-orin-nanoedge-aivla-modelsroboticsmultimodal-modelshugging-faceopen-modelsai-infrastructure
NVIDIA Demonstrates Gemma 4 VLA Model Running on Jetson Orin Nano Super — Intelligence — Agentic Developer Tools Radar · Signal