JAKARTA Nvidia, opened a new tab announcing on Tuesday 9 September that it will launch a new artificial intelligence (AI) chip by the end of next year. This chip will be specially designed to handle complex functions such as video and software creation.

The new chip, named "Rubin CPX", will be built with Nvidia's latest generation Rubin architecture the successor to the "Blackwell" technology which is the company's first step in providing a larger processing system.

Along with the increasingly sophisticated AI system, which handles heavy tasks such as "vide coding" (AI-assisted code making) and video creation, the industry's need for data processing is increasing.

Nvidia stated that the AI model could take up to 1 million tokens to process an hour of video content a challenging achievement for traditional GPUs. Tokenmerujuk on the data unit processed by an AI model.

To overcome this, Nvidia will integrate various steps from long processing sequences, such as decodingvideo, encoding, and inference when the AI model generates output into its new chip.

The company said that an investment of $100 million in this new system could help generate token revenues of $5 billion, as Wall Street's growing focus on returns from hundreds of billions of dollars in investment into AI hardware.

The race to develop the most advanced AI system has made Nvidia the most valuable company in the world, controlling the dominant AI chip market share with its most expensive and most expensive processors.


The English, Chinese, Japanese, Arabic, and French versions are automatically generated by the AI. So there may still be inaccuracies in translating, please always see Indonesian as our main language. (system supported by DigitalSiber.id)