Intel, with its predecessor chip, had touted its performance as close to Nvidia’s top chip of the time, the H100. It claimed that with Gaudi 3, it’s pointing to large-language model (LLM) performance of 40 percent faster training time for the GPT-3 175B large language model versus the H100. But looming in the background is Nvidia’s next GPU, the Blackwell B200.
Intel, with its predecessor chip, had touted its performance as close to Nvidia’s top chip of the time, the H100. It claimed that with Gaudi 3, it’s pointing to large-language model (LLM) performance of 40 percent faster training time for the GPT-3 175B large language model versus the H100. But looming in the background is Nvidia’s next GPU, the Blackwell B200.