OpenAI and Broadcom have announced Jalapeno, a custom AI inference accelerator designed for large language model (LLM) workloads, marking OpenAI's first proprietary processor and the start of a planned multi-generation compute platform. The companies said the chip has been architected specifically for LLM inference rather than adapted from existing AI accelerator designs. OpenAI developed the processor with Broadcom and systems integrator Celestica, combining chip design, networking, rack integration and power delivery.