OpenAI touts Jalapeño chip for faster AI inference
By The BriefBrief
OpenAI says its Jalapeño AI chip completes inference tasks more efficiently and delivers faster responses than competing AI systems. In tests using InferenceX, OpenAI reports Jalapeño achieved 1.5 to 1.9 times more work per watt and 1.7 to 3.6 times lower end-to-end latency than Nvidia GB200/GB300 across several models.
WHO:OpenAI hardware vice president Richard Ho
WHAT:OpenAI claims its Jalapeño ASIC outperforms competing superchips on AI inference efficiency and latency benchmarks.
