NVIDIA dominates AI benchmark rankings, significantly outperforming competing chips.

Nvidia's general-purpose GPU chips have swept one of the most popular benchmarks for measuring chip performance in artificial intelligence, with a new focus on generative AI applications such as large language models (LLMs). Systems assembled by SuperMicro, Hewlett Packard Enterprise, Lenovo, and others—packed with as many as eight Nvidia chips—took most of the top honors in the MLPerf benchmark test organized by the MLCommons, an industry consortium on Wednesday.

The test measures how fast machines can produce tokens, process queries, or output samples of data—a process known as AI inference. This represents the fifth installment of the prediction-making benchmark. The new version of the MLPerf tests simulates how fast a chatbot might respond to prompts, emphasizing real-world applications of generative AI systems.