Insights
Blog
Insights, technology and updates from the REBOTNIX team.
AI on NVIDIA Jetson with a smaller runtime, not a smaller model
A complete detection pipeline in 501 MB at a measured 104 fps end to end on Jetson Orin NX, with unchanged model output. Four models in 725 MB instead of 2.16 GB. Why the model is rarely where the memory goes.

REBOTNIX to Support New NVIDIA Jetson T3000 and T2000 Modules Across Its GUSTAV Edge AI Platform
NVIDIA Preferred Partner REBOTNIX extends its Jetson Thor roadmap. Blackwell class inference in compact form factors, one software stack, and a frictionless migration path. Evaluation in emulation mode starts this month.

REBOTNIX makes smart city cameras with agentic reasoning capabilities for faster city-level decision-making with NVIDIA
Up to seven parallel Vision Transformer models, AGI orchestrators and NVIDIA Jetson Thor. How REBOTNIX automatically detects municipal infrastructure events and routes them into operational processes.