Nvidia announced its next-generation Jetson Orin Nano 2 robotics computing platform on Tuesday, U.S. Eastern Time, as it expands AI from data centers into physical devices such as robots and drones. The compact platform targets cost-, power- and latency-sensitive edge AI applications and delivers twice the inference performance of the Jetson Orin Nano Super. It includes an 8-core processor, 8GB of memory and up to 78 TOPS, or 78 trillion operations per second. Upgraded tensor cores and higher memory bandwidth support memory-efficient large language and vision-language models including Cosmos, Nemotron, Gemma 4 and Qwen 3. In 15-watt mode, it matches the previous generation's performance while consuming 40% less power. Nvidia says the platform can support local multimodal processing and real-time decision-making for home robots, vision AI systems, delivery drones and inspection drones. Wing plans to evaluate it after deploying the Jetson Orin Nano Super and Nvidia's software stack in its delivery drones, while Matic plans to use it for natural-language interaction, gesture recognition, spatial mapping and semantic understanding. Nvidia also demonstrated the platform running Reachy Mini while simultaneously handling language, speech and real-time vision AI models. More than 3 million developers are building on Nvidia's robotics technology stack, and partners including AverMedia, ADLINK Technology, Advantech, Aetina, Antmicro, Aptiv, Connect Tech and Seeed Studio are developing related hardware and software. Cognex, South Korea's Doosan Bobcat and Matic are among the first companies adopting or exploring the platform. The Jetson Orin Nano 2 module and developer kit are expected to launch in the first half of 2027, so its near-term financial contribution remains uncertain. Nvidia is positioning the product as part of a broader physical AI ecosystem spanning models, software, computing platforms and robot endpoints.