Good news for AI developers and hobbyists: Nvidia has made it much less expensive to build robots, drones, smart cameras, and other AI-powered gadgets. The company's new Jetson Oren Nano Super Announced Tuesday Available now, it has more processing power than its predecessor while costing half the price at $249.

The palm-sized computer delivers a 70% performance boost, reaching 67 trillion operations per second for AI tasks. This is a big leap from previous models, especially for powering things like chatbots, computer vision, and robotics applications.

“This is a brand new Jetson Nano supercar. Nearly 70 trillion operations per second, 25 watts and $249,” Nvidia CEO Jensen Huang said in an official video from his kitchen. “It runs everything HGX does, even They run LLMs.”

Memory bandwidth also got a major upgrade, increasing to 102GB/s, 50% faster than the previous generation Jetson. This improvement means the device can handle more complex AI models and process data from up to four cameras simultaneously.

The device comes with Nvidia's Ampere architecture GPU and a hexa-core ARM processor, allowing it to run multiple AI applications simultaneously. This gives developers the ability to work with more diverse competencies, such as building small models of robots capable of doing things like environment mapping, object recognition, and voice commands with low processing power.

Current Jetson Orin Nano owners aren't left out in the cold either. Nvidia launches Software updates To increase the efficiency of old artificial intelligence processors.

the numbers Behind Nvidia's new Jetson Orin Nano Super tells an interesting story. With just 1,024 CUDA cores, it seems modest compared to the RTX 2060's 1,920, the RTX 3060's 3,584, or the RTX 4060's 3,072. But the raw core count doesn't tell the whole story.

While gaming GPUs like the RTX series consume between 115 and 170 watts of power, the Jetson only consumes between 7 and 25 watts. That's about one-seventh of the power consumption of the RTX 4060, which is the most efficient of the bunch.

Memory bandwidth numbers paint a similar picture. The Jetson's 102 GB/s speed may seem disappointing compared to 300+ GB/s RTX cards, but it's specifically optimized for edge AI workloads. Process data efficiently It matters more than raw productivity.

However, the real magic happens in the performance of the AI. The device produces 67 TOPS (trillion operations per second) for AI tasks, a number that is difficult to compare directly with the TFLOPS of RTX cards since they measure different types of operations.

But in Practical termsJetson can handle tasks like running local AI chatbots, processing multiple camera feeds, and controlling bots — all at once on a power budget that can barely power a gaming GPU's cooling fan, essentially being on par with The RTX 2060 is at a fraction of the cost and a fraction of the power consumption.

8GB of shared memory may seem low, but it means it's more capable than the regular RTX 2060 when it comes to running local AI models like flow or Stable spread Which could trigger an “out of memory” error on these GPUs, or offload some of the work to regular RAM, reducing inference time — which is essentially the AI’s thinking process.

Jetson Orin Nano Super also supports many features Small and large linguistic modelsincluding those with up to 8 billion parameters, such as the Llama 3.1 model. It can generate codes at a rate of approximately 18-20 per second when using a quantum version of these models. A little slow, but still good enough for some native applications. However, it represents an improvement over the previous generation of Jetson AI hardware models.

Given its price and features, the Jetson Orin Nano Super is designed primarily for prototyping and small-scale applications. For power users, businesses, or applications that require extensive computational resources, the device's capabilities may seem limited compared to high-end systems that cost much more and require more power.

Modified by Andrew Hayward