OpenAI, Broadcom Introduce New Intelligence Processor Built for LLM Inference

 OpenAI and Broadcom have introduced Jalapeño, OpenAI's first Intelligence Processor, an accelerator built around OpenAI's approach to LLM inference and the first chip in a multi-generation compute platform the two companies are developing together. The goal is to make advanced AI faster, more reliable, and more accessible.

The chip was handed over to OpenAI CEO Sam Altman and President Greg Brockman by Broadcom's President and CEO Hock Tan and Semiconductor Solutions President Charlie Kawwas, marking a notable step in OpenAI's push to build out the full technology stack behind its models and products.

OpenAI designed Jalapeño from the ground up, drawing on its understanding of LLM fundamentals and its roadmap across models, kernels, serving systems, and product needs. Broadcom and Celestica supported the effort by helping industrialize the platform through chip implementation, board and rack system integration, high performance networking, and scalable production. The chip is built with the flexibility to support LLMs broadly, shaped by OpenAI's insight into inference requirements across current and future AI models industry wide. Engineering samples are already running machine learning workloads in the lab at production-target frequency and power, including GPT-5.3-Codex-Spark.