OpenAI claims its new AI chip beats Nvidia on inference benchmarks

OpenAI unveiled its Jalapeño chip, an ASIC designed for AI inference, claiming it offers both lower latency and higher throughput than competing systems. In benchmark tests, the chip delivered 1.5 to 1.9 times more work per watt and 1.7 to 3.6 times lower end-to-end latency compared to Nvidia's GB200 and GB300 superchips. The chip was developed in partnership with Broadcom.