OpenAI and Broadcom announce chip designed for LLM inference at scale

OpenAI and Broadcom unveiled Jalapeño, OpenAI's first Intelligence Processor designed specifically for LLM inference , with early testing showing substantially better performance per watt than current state-of-the-art . The chip is designed with flexibility to work with all LLMs and will be deployed at gigawatt scale with Microsoft and other partners beginning in 2026 . The chip was co-developed from initial design to manufacturing tape-out in just nine months , marking a key step in OpenAI's vertical integration strategy.

Source: Ars Technica | 2026-06-24