OpenAI's Jalapeño Chip Outperforms Nvidia Blackwell on Independent Inference Benchmark

At Hot Chips on August 25, 2026, OpenAI showed first results for Jalapeño, its custom inference chip built with Broadcom. On SemiAnalysis's independent InferenceX benchmark, Jalapeño produced more tokens per user and higher throughput per kilowatt than an Nvidia Blackwell system. Limited deployment is planned for late 2026, with broader rollout in 2027.