Woofun AI reports that OpenAI has commenced production of its proprietary 'Jalapeño' application-specific integrated circuit, specifically engineered for large-language model inference tasks. Developed in collaboration with Broadcom and Celestica, the chip reached manufacturing readiness in just nine months, significantly faster than the industry standard of 18 to 36 months for high-performance ASICs.

The deployment marks a strategic shift toward vertical integration, aiming to control the entire intelligence creation pipeline from models to hardware. By utilizing GPT models to optimize chip design, OpenAI seeks to reduce inference costs and improve system throughput. Unlike training-focused GPUs, Jalapeño addresses ongoing operational expenses, allowing OpenAI to lower token production costs and enhance efficiency across ChatGPT and Codex services without relying solely on external hardware suppliers.