20 Hrs
المصدر:
مينا تك
مينا تك
جاهز للتشغيل
جاهز للتشغيل
OpenAI has launched their first dedicated inference processor called Jalapeño, in collaboration with Broadcom, aimed at improving performance efficiency and reducing energy consumption for AI models, especially in inference operations used for deploying and interacting with pre-trained AI models. The chip is still in the testing phase, and initial results have shown a significant improvement in efficiency compared to current market solutions, with the company moving towards reducing reliance on Nvidia graphics processing units.
تنويه: هذا ملخص تم إنشاؤه بواسطة الذكاء الاصطناعي
comments.heading