Nvidia unveils new GPU designed for long-context inference

Core Insights - Nvidia announced the Rubin CPX GPU, designed for context windows larger than 1 million tokens, at the AI Infrastructure Summit [1] - The Rubin CPX is part of Nvidia's upcoming Rubin series and is optimized for processing large sequences of context, enhancing performance in long-context tasks such as video generation and software development [1] - Nvidia reported $41.1 billion in data center sales in its most recent quarter, reflecting the company's strong financial performance [2] - The Rubin CPX is expected to be available by the end of 2026 [2]