Search results for Hopper
Industry
Nvidia shifts AI chip battleground from specs to end-to-end efficiency
Nvidia said it will shift competition in AI semiconductors from chip specifications to end-to-end efficiency across pre-training, post-training, inference and agents. It also disclosed measured results showing Blackwell-based GPUs deliver 55 times faster mixture-of-experts inference than the prior Hopper generation. Nvidia highlighted efficiency gains from a new numeric format and software advances, and introduced curriculum-based post-training results for its Nemotron 3 Nano model. It also said Korean partners are participating and unveiled a Korean-focused synthetic dataset.
AI & Enterprise
Fujitsu unveils document automation AI cutting legacy code understanding time by 97 percent
Fujitsu has launched a generative AI service that analyses legacy source code, including COBOL, and automatically produces design documents. The company said it aims to cut code understanding and documentation work from hours to minutes and reduce reliance on specialist staff. Fujitsu said it cut time needed to understand complex code by about 97 percent and improved document comprehensiveness and readability using enhanced retrieval linking large codebases via a knowledge graph.
AI & Enterprise
China\'s Illuvatar CoreX unveils multi-year GPU roadmap, targets global market
Chinese AI semiconductor startup Illuvatar CoreX Semiconductor unveiled a multi-year GPU architecture roadmap and introduced an edge computing product line. It said its first architecture, Tianshu, outperformed Nvidia\'s Hopper platform, and said later designs target Blackwell and Rubin. The company also introduced TY-series edge products and said TY1000 surpassed Nvidia\'s AGX Orin in tests. It listed in Hong Kong on Jan. 8 and reported rising revenue alongside widening losses.
