NVIDIA Launches Jetson Thor T3000 and T2000 Modules for Mass-Market Robotics and Edge AI

NVIDIA introduced the Jetson and IGX T3000 modules, delivering 865 FP4 teraflops of AI compute in a form factor roughly half the size and power of the prior T5000. The T3000 combines a Blackwell GPU, an eight-core Neoverse Arm CPU, 32GB of LPDDR5X memory, 273GB/s memory bandwidth, and 25 GbE connectivity. The IGX T3000 variant adds integrated functional safety and runs NVIDIA Halos for Robotics, a full-stack safety system for robots operating alongside humans. Despite the smaller footprint, NVIDIA says the T3000 achieves inference performance similar to the T5000 on multimodal workloads including large language models, vision-language models, vision-language-action models, and world foundation models. The company also unveiled the Jetson T2000, an entry point for the Thor architecture with 400 FP4 teraflops and 16GB of memory, targeting visual AI agents, autonomous mobile robots, and industrial manipulators. NVIDIA said the new modules are being adopted by 1X, Agile Robots, Amazon Robotics, Boston Dynamics, FANUC, Hitachi, and Techman Robot for humanoid and industrial robotics deployments.