Connect with us


Google reveals that its latest A.I. supercomputer beats Nvidia



Google TPU 4

You must be aware that Google is a company that competes with a lot of industry giants as it operates in multiple industries. For example, Google competes with Microsoft and Amazon for its cloud platform while it competes with phone manufacturers like Apple and others in the smartphone industry. Similarly, it also builds supercomputers so it is in direct competition with GPU makers like Nvidia and AMD as well. Recently, Google announced that it has built an AI Supercomputer which is faster and more efficient than competing Nvidia systems.

Google’s AI chips are called Tensor Processing Units, and they are competing with Nvidia’s AI chips which are currently dominating the market with 90% deployment. Google’s researchers say that its TPU v4 is “1.2x–1.7x faster and uses 1.3x–1.9x less power than the Nvidia A100,” They added that “The performance, scalability, and availability make TPU v4 supercomputers the workhorses of large language models,” The researcher from Google wrote that “Today’s MLPerf 3.0 highlights Hopper delivering 4x more performance than A100,”. “The next level of Generative AI requires new AI infrastructure to train Large Language Models with great energy efficiency”. Google revealed that MidJourney was trained on its TPU chips.

Google is considered to be a pioneer in the AI and ML field but the fact that it has failed to commercialize this technology that has already been built and also the fact that OpenAI’s ChatGPT was deployed earlier than Google’s Bard and also integrated into Microsoft’s Bing gives everyone the perception that Google has slowed down and this is a real concern among the investors who are only interested in the results and not about how accurate the technology is. Having said that, OpenAI’s GPT-4 is no slouch as it has already beaten the Google Bard in a lot of tests and seems to be the better platform for AI chatbots right now.

One thing that Microsoft and OpenAI should get credit for is how fast they were able to deploy the ChatGPT into the hands of people which meant that everyone started beta testing it and this is the reason why it has improved to such an extent right now.

Click to comment

Leave a Reply

Your email address will not be published. Required fields are marked *