ThinkingNews
Topics

Topic

Broadcom

4 written stories · 6 tracked

Tech4 outlets

A detailed look at Jalapeño, OpenAI's ASIC developed with Broadcom in 16 months, which beat Nvidia, AMD, and Google chips on multiple top open-source models (SemiAnalysis)

OpenAI’s Jalapeño ASIC, developed with Broadcom in about 16 months, is designed for fast inference at scale and has been benchmarked against leading competitors. Tests show it delivers roughly 1.5-to-1.9 times more AI work per watt and 1.7-to-3.6 times lower latency than Nvidia chips, and it outperforms the GB300 and comparable Nvidia, AMD and Google chips on several top open-source models.

The Verge·TechCrunch·TechMeme·The Next Web
Tech5 outletsover 5h

Apple says it reached a $30B+ deal with Broadcom to produce 15B chips on US soil in the next five years, its largest pledge as part of its $600B US investments (Rolfe Winkler/Wall Street Journal)

Apple has agreed to a $30 billion-plus contract with Broadcom to manufacture 15 billion chips in the United States over the next five years. The deal is part of Apple’s broader $600 billion U.S. investment plan. The chips will be made in America and will include new wireless components.

TechMeme·Hacker News·The Next Web·Engadget·TechCrunch
Tech7 outletsover 9h

OpenAI and Broadcom unveil Jalapeño, an LLM-optimized inference chip developed from design to manufacturing tape-out in nine months, aided by OpenAI's models (OpenAI)

OpenAI and Broadcom have unveiled Jalapeño, a custom inference chip optimized for large-language-model (LLM) workloads. The chip was designed, manufactured, and tape-out completed in nine months, a process accelerated by using OpenAI’s own models. Jalapeño is intended for large-scale LLM inference and represents the first AI processor jointly developed by the two companies.

TechMeme·The Next Web·The Verge·TechCrunch·VentureBeat+2