LaunchDevelopersAugust 25, 2026

OpenAI's Jalapeño chip beats Nvidia on inference benchmarks

OpenAI says its Jalapeño ASIC, made with Broadcom, delivered 1.5–1.9x more AI work per watt and 1.7–3.6x lower latency than Nvidia's GB200/GB300 across GPT-OSS 120B, DeepSeek R1, and Kimi K2.5 1T. Deployment starts in small volumes by end of 2026, ramping into 2027.

Featured · Richard Ho

1 source

More stories today

Open the live feed