OpenAI’s Jalapeño chip is built for fast inference at scale, benchmarks show

August 25, 2026

Tested on Semianalysis’s InferenceX benchmark, Jalapeño registered both more tokens per user and more throughput per kilowatt than the currently available state-of-the art.