Search

Showing top 2 results for "Jalapeño performance claims"

OpenAI's upcoming Jalapeño AI chip outperforms NVIDIA GB300 in inference tests

… For highly interactive workloads, OpenAI claims a 2.1x to 4.1x performance advantage. For example, while running DeepSeek R1, Jalapeño delivered 19,641 mixed tokens per second per kilowatt, compared to 11,781 for the NVIDIA GB300. …

Aug 26, 2026 · Pradeep Viswanathan

OpenAI and Broadcom unveil Jalapeño, a new AI chip built for LLM inference

… OpenAI pointed out that Jalapeño went from initial design to manufacturing tape-out in just nine months, which it claims is the fastest ASIC development cycle achieved for a high-performance advanced semiconductor. …

Jun 24, 2026 · Pradeep Viswanathan

Followed topics

OpenAI's upcoming Jalapeño AI chip outperforms NVIDIA GB300 in inference tests

OpenAI and Broadcom unveil Jalapeño, a new AI chip built for LLM inference