Artificial intelligence is becoming increasingly widespread. Companies are constantly developing new AI technologies to compete with or integrate with OpenAI’s ChatGPT. Cisco is the latest company to join this race, but not in the way you might think. Instead of developing new AI software, Cisco is launching new AI networking chips. These chips aim to rival the offerings from Broadcom and Marvell. Here are the details…

New Cisco AI Networking Chips Aim to Speed Up AI and Machine Learning Tasks

Cisco has launched new AI networking chips that are designed to compete with the offerings from Broadcom and Marvell. These chips are part of Cisco’s SiliconOne series and are being tested by five of the six major cloud providers, including Amazon Web Services, Microsoft Azure, and Google Cloud.

Cisco AI chip

The new chips are said to be the most powerful networking chips in the market for AI/ML workloads. They can connect up to 32,000 GPUs together, which can help to speed up AI and machine learning tasks. Cisco also says that the chips can help to reduce the number of switches needed in a network, which can further improve performance.

As the use of AI grows, so does the demand for AI networking chips. These chips help companies build and deploy AI applications more quickly and efficiently. AI applications are becoming more complex and demanding, which means that they need high-performance chips. Cisco is meeting the growing demand for AI networking chips.

The company’s new chips are powerful and energy-efficient, which is important for AI applications. This could help companies save money on their energy bills. Cisco is a major player in the networking industry, and its new chips could help the company to gain market share in this growing market.

RELATED:

(via)