A compact robotics module that doubles inference performance of its predecessor while cutting power use, aimed at home robots, drones, and vision systems running LLMs and vision-language models at the edge.
Jetson Orin Nano 2 delivers 78 trillion operations per second of AI compute, with 8 GB memory and an eight-core Arm CPU, offering about twice the inference performance of the prior Jetson Orin Nano Super in the same form factor.
The module reduces power consumption: Nvidia reports it can use 40% less power than the previous model when providing the same performance in 15-watt mode.
# What Nvidia announced
Nvidia introduced the Jetson Orin Nano 2, a compact robotics computer intended for entry-level physical AI use cases: robots, drones, vision systems and other devices that need real-time perception, reasoning and decision-making at the edge. The module delivers 78 trillion operations per second (TOPS) of AI compute, includes 8 GB of memory and an eight-core Arm CPU.
# Performance and efficiency
# Running generative and multimodal models on-device
Jetson Orin Nano 2 is built to run large language models (LLMs) and vision-language models locally. Nvidia highlights that smaller, more efficient frontier models make it feasible to perform image-and-language processing and respond to surroundings without continuous cloud access. Nvidia specifically cited model families that developers can use for memory-efficient edge inference: Cosmos, Nemotron, Gemma 4 and Qwen 3.
The platform integrates with Nvidia's software stack, Jetson agent skills and its broader AI ecosystem to support on-device generative and multimodal use cases.
# Early adopters and targeted applications
Nvidia named several companies already evaluating or adopting the platform:
# Hardware ecosystem and availability
A wide group of Nvidia partners is preparing carrier boards, complete systems, customized AI software and reference designs for Orin Nano 2. Partners listed include AAEON, ADLINK, Advantech, Aetina, Antmicro, Aptiv, Auvidea, Seeed Studio and others. These third-party products are intended to shorten development cycles and accelerate production readiness for Orin Nano 2-based systems.
Nvidia says more than 3 million developers are building on its robotics stack. The Jetson Orin Nano 2 module and a developer kit are expected to become available in the first half of 2027.
# What this means for edge robotics
By increasing inference performance and lowering power draw in the same compact form factor, Orin Nano 2 targets devices where size, battery life and latency matter. The platform's explicit support for memory-efficient LLMs and vision-language models signals Nvidia's intent to push generative and multimodal AI into smaller physical machines that must act in real time with limited cloud connectivity.
Nvidia has released the Jetson Orin Nano 2, a new robotics computer for entry-level edge AI. The company says the new computer will enable millions of developers worldwide to build robots, delivery and inspection drones, and vision AI systems for frontier physical AI applications. Jetson Orin Nano 2 delivers 2x the inf
D-Robotics RDK S100P development kit is a robotics SBC built around an S100P-based module providing an alternative to NVIDIA Jetson Orin NX 16GB with a 6-core Cortex-A78AE application cluster, a 4-core Cortex-R52+ MCU domain, a Nash BPU rated at 128 TOPS INT8, a Mali-G78AE GPU, and 24 GB of LPDDR5. The four R52+ cores
Built for robotics development, drone vision, and edge AI prototyping on the NVIDIA Jetson Orin Nano, this camera pairs the Onsemi AR2020 sensor with VCM autofocus over MIPI CSI-2, giving developers genuine 20MP detail and consistent focus validated specifically for the Orin Nano's compact edge AI compute platform.
Back in November 2025, we wrote about the Brainchip AKD1500 PCIe/SPI Edge AI co-processor designed for low-power systems and delivering up to 800 GOPS at just 300 mW. In recent months, the company has launched several hardware platforms based on the ADK1500. At the end of July, Brainchip introduced the AKD1500 M.2 B+M
Nvidia has released Isaac ROS 5.0, introducing AI agent capabilities designed to help developers build, customize and deploy robotics applications more quickly using the open-source Robot Operating System ecosystem. Released at ROSCon in Toronto, Canada, Isaac ROS 5.0 adds new agentic development workflows alongside ex
“Computing power” and “compute” are becoming key features of the AI boom, both for tech enthusiasts and ordinary people alike.... The post What Is A Computer Cluster Or Cluster Computing? appeared first on TechRound .
Loading more related stories...
Open the app view to save this story, compare related coverage, and continue from the same source.