Rumors say that Microsoft is about to integrate a neural processor (Neural Processing Unit, NPU) into all future Surface devices. This component will significantly speed up calculations related to artificial intelligence and machine learning.
Microsoft is increasing usage of artificial intelligence (AI) in their hardware products. The company is incorporating AI into their devices to enhance the user experience, automate tasks, and improve overall performance.
Currently, the Neural Processing Unit (NPU) is only found in Qualcomm's desktop processors, and the Surface Pro 9 is the first device in the line to feature this AI accelerator. Intel and AMD are also working on integrating NPUs into their products.
The Athena NPU
In other news, Microsoft is developing its own Athena processor to accelerate artificial intelligence tasks. The project exists since 2019. The company expects that this development will help save money and eliminate dependence on NVIDIA solutions.
Now some Microsoft and OpenAI employees have already received chips codenamed Athena to test large language models such as GPT-4. As you may remember, it is the language model that underlies the recently launched Bing chatbot.
Analysts estimate that around 30,000 NVIDIA GPUs are required to meet the demand for a service like ChatGPT. Given that the cost of the A1000 graphics adapter ranges from 10 to 15 thousand dollars, running a conditional ChatGPT will cost about $300 million.
Support us
Winaero greatly relies on your support. You can help the site keep bringing you interesting and useful content and software by using these options:
If you like this article, please share it using the buttons below. It won't take a lot from you, but it will help us grow. Thanks for your support!
Advertisеment
There must be a downside to this !