OpenAI Jalapeño delivers higher throughput per kilowatt and lower latency than rivals on GPT-OSS 120B, with deployment set for year-end.

DeepSeek develops custom inference chip after crashing Nvidia $600 billion market cap, raising $7 billion amid US export controls and Huawei Ascend competition.

OpenAI and Broadcom launch Jalapeño, a custom AI chip built for LLM inference to raise performance, efficiency and scale.