OpenAI and Broadcom today unveiled Jalapeño, OpenAI's first Intelligence Processor designed specifically for LLM inference, marking a key step in OpenAI's full-stack infrastructure strategy. The chip was delivered to OpenAI CEO Sam Altman and President Greg Brockman by Broadcom leadership.
Jalapeño was developed from initial design to manufacturing tape-out in nine months, representing the fastest ASIC development cycle in high-performance semiconductors according to both companies. Early testing shows the chip delivers substantially better performance per watt than current state-of-the-art systems, with engineering samples already running production workloads including GPT-5.3-Codex-Spark. The platform combines OpenAI-designed accelerators with Broadcom silicon and networking technologies, with initial deployment targeted for end of 2026.