A detailed look at Jalapeño, OpenAI's ASIC developed with Broadcom in 16 months, which beat Nvidia, AMD, and Google chips on multiple top open-source models (SemiAnalysis)

OpenAI has developed a custom artificial intelligence chip called Jalapeño in collaboration with Broadcom. The chip was created in 16 months and has shown to outperform Nvidia, AMD, and Google chips on multiple top open-source models. According to OpenAI, Jalapeño delivers 1.5 to 1.9 times more AI work per watt of power and 1.7 to 3.6 times lower latency compared to rival chips.

The chip was tested using InferenceX, an independent benchmark from SemiAnalysis, on several large language models including GPT-OSS 120B, DeepSeek R1, and Kimi K2.5. OpenAI's head of hardware, Richard Ho, stated that Jalapeño can serve more AI work per unit of power while returning responses more quickly. The company plans to deploy Jalapeño at the end of 2026 in small volumes, with more significant deployment expected in 2027.

Jalapeño was designed specifically for inference, the computing stage during which a trained AI model generates answers. OpenAI's own AI models assisted in the development process of the chip, which is part of the company's push to gain greater control over the computing infrastructure powering ChatGPT and its other AI services.

Brief written by urgent.news from Techmeme, Digital Trends, Business Insider, Arabian Post, TechCrunch — 5 reports on this story. Machine-written — may contain errors; check the original before relying on it.